Physical ai-based humanoid robot polarization microscope automatic manipulation and central hub-linked mineral identification support system

Abstract

The objective of the present invention, which aims to solve the aforementioned conventional problems, is to provide a physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system capable of reducing the time required and observation deviations caused by manual user operation by automatically performing microscope operations required in the mineral identification process—such as rotating the polarizer, inserting and removing the analyzer, rotating the stage, adjusting the focus, and changing the magnification—through a human robot. To achieve the above objective, the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention comprises, in a physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system for identifying the type of mineral contained in a mineral sample, a polarizing microscope device for observing the mineral sample; a human robot configured to physically operate at least one of a polarizing plate rotation module, an analyzer insertion module, a stage rotation module, a focus adjustment module, and a magnification change module of the polarizing microscope device; and a central hub system that integrally manages polarizing microscope operation learning data and control algorithms corresponding to a plurality of polarizing microscope models and provides polarizing microscope operation learning data and control algorithms corresponding to the model of the polarizing microscope device to the human robot, wherein the central hub system comprises a data storage unit that stores polarizing microscope operation learning data and control algorithms for a polarizing microscope operation procedure performed in the mineral identification process of an experienced researcher, and polarizing microscope operation learning data corresponding to the model of the polarizing microscope device It includes a transmission unit that transmits a control algorithm to the human robot, and the human robot is configured to recognize the position or shape of the operation part of the polarizing microscope device and automatically perform operations related to the mineral identification process for the polarizing microscope device based on the polarizing microscope operation learning data and control algorithm received from the central hub system.

Description

Technology Field

[0001] The present invention relates to a physical AI-based human-robot polarizing microscope automatic operation and central hub-linked mineral identification support system, and more specifically, to a physical AI-based human-robot polarizing microscope automatic operation and central hub-linked mineral identification support system that automates the operation of a polarizing microscope using a human-robot-based physical AI and supports mineral identification by linking with a central hub-based learning data and algorithm provision system. Background Technology

[0002] In the field of mineral analysis, mineral identification techniques using polarizing microscopes are widely utilized to determine the types of minerals contained in rock or mineral samples. Generally, during the identification process, mineral thin sections or powdered samples are placed on a polarizing microscope, and the researcher directly operates the microscope to observe the optical properties of the minerals. In this process, the researcher repeatedly performs various microscopic operations—such as rotating the polarizer, inserting and removing the analyzer, rotating the stage, changing magnification, and adjusting the focus—to comprehensively assess the mineral's interference color, extinction characteristics, birefringence properties, twinning structure, and crystal morphology, thereby identifying the type of mineral.

[0003] In particular, mineral identification based on polarizing microscopy involves a process that goes beyond simply capturing and analyzing a single image; it includes verifying the mineral's optical response stepwise under different polarization conditions and observation angles. For instance, researchers can observe changes in the mineral's interference colors by repeatedly switching between crossed and open nicol states, and they may rotate the stage to a specific angle to observe extinction characteristics occurring at particular angles. Additionally, processes such as finely adjusting the focal position or changing the magnification may be performed to clearly identify the mineral's crystal structure or grain boundaries.

[0004] For a long time, such mineral identification processes have relied on the experience and judgment of skilled researchers, and in actual research settings, analyses are mostly performed by mineralogists or experts in microscopic analysis who directly operate polarizing microscopes. Consequently, even for the same mineral sample, analysis results may vary depending on the researcher's proficiency, observational experience, and operating method; furthermore, there is a problem in accurately identifying the type of mineral when sufficient observation conditions for a specific mineral are not secured.

[0005] Meanwhile, with the recent advancement of artificial intelligence technology, various attempts are being made to automatically identify or classify minerals based on polarizing microscope images. For example, methods can be utilized to automatically classify specific minerals by inputting mineral images captured by a polarizing microscope into a deep learning-based AI model, or to extract mineral textures and optical features and compare them with a database. Additionally, some technologies are conducting research to improve mineral classification accuracy by utilizing large amounts of mineral image data as training data.

[0006] However, conventional AI-based mineral analysis technologies mostly focus solely on the final images captured by polarizing microscopes, which limits their ability to adequately consider the microscope manipulation procedures required during the actual mineral identification process. In other words, while appropriate polarization, observation angle, focus, and magnification conditions must be secured to clearly reveal mineral characteristics, conventional technologies suffer from the problem that the process of establishing these observation conditions relies entirely on the user.

[0007] In particular, during the mineral identification process, since the characteristics of specific minerals often become clearly apparent only under specific polarization conditions or rotation angles, the accuracy of AI analysis results can be significantly degraded if researchers capture images without performing sufficient observation procedures. For example, if the extinction characteristics of a specific mineral are not observed under appropriate stage rotation conditions, or if images are captured without securing polarization conditions where interference colors are clearly revealed, there is a possibility that the AI ​​model will fail to accurately distinguish between minerals with similar optical properties.

[0008] Furthermore, since the mineral identification process based on polarizing microscopes involves repetitive manual operations, analysis times can be significantly prolonged, which can lead to increased researcher fatigue in research environments where a large number of samples must be analyzed. In particular, because researchers must repeatedly manipulate polarizers, rotate the stage, and adjust the focus, operational accuracy may deteriorate during prolonged analysis, which can result in lower reproducibility of the analysis results.

[0009] Furthermore, polarizing microscope devices often differ in operation methods, interface structures, knob positions, and control mechanisms depending on the manufacturer or model. Since operating structures can vary—for example, some devices use mechanical rotary knobs while others utilize electronic control interfaces—there is a problem in that automation technology optimized for a specific microscope model is difficult to apply directly to other devices. Consequently, there are limitations in establishing an integrated automation system in environments where research institutions use different models of polarizing microscopes.

[0010] Furthermore, since existing automation technologies are often implemented in a manner dependent on specific devices, inefficiencies may arise as separate control programs or interfaces must be redeveloped whenever a new microscope device is introduced. This can limit the scalability of microscope automation technology and presents challenges in its universal application across diverse research environments.

[0011] Meanwhile, since the microscopic operation and analysis data accumulated during the mineral analysis process are mostly managed independently within individual research institutions, there is a problem in that the identification procedures or analytical know-how of skilled researchers are not systematically shared. For example, even if an effective observation procedure for a specific mineral is established at a particular research institution, such procedures are often not efficiently transferred to other institutions or researchers, which can lead to persistent variations in analysis among researchers.

[0012] Furthermore, since most existing AI-based mineral analysis systems operate based on static training data, there is a problem in that it is difficult to continuously improve analysis algorithms by reflecting new mineral data or microscopic observation conditions in real time, even when they are added. In particular, there is a possibility that analysis accuracy may deteriorate if differences in image characteristics or operating environments acquired from different polarizing microscope models are not sufficiently reflected.

[0013] Therefore, there is a need for a new type of mineral analysis technology that can systematically manage and automate the microscope operation procedures performed during the mineral identification process, flexibly adapt to polarizing microscope environments of different manufacturers and models, and continuously learn and expand the analysis procedures and operation data of skilled researchers. Prior art literature

[0014] Korean Registered Patent Publication No. 10-1707440 (Published on Feb. 20, 2017) The problem to be solved

[0015] The objective of the present invention, which aims to solve the aforementioned conventional problems, is to provide a physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system capable of reducing the time required and observation deviations caused by manual user operation by automatically performing microscope operations required in the mineral identification process—such as rotating the polarizer, inserting and removing the analyzer, rotating the stage, adjusting the focus, and changing the magnification—through a human robot.

[0016] Furthermore, the objective of the present invention is to provide a physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system capable of stably forming observation conditions suitable for mineral identification and securing images suitable for artificial intelligence analysis by establishing the polarizing microscope operation procedure performed by a skilled researcher during the mineral identification process as learning data and control algorithms, and providing this to a human robot.

[0017] Furthermore, the objective of the present invention is to provide a physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system that enables a human robot to recognize and respond to the position and shape of the control unit even for polarizing microscopes of different manufacturers or models, and provides a mineral identification automation platform scalable in various research environments by integrating and managing learning data and algorithms for this purpose from a central hub. means of solving the problem

[0018] To achieve the above objective, the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention comprises, in a physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system for identifying the type of mineral contained in a mineral sample, a polarizing microscope device for observing the mineral sample; a human robot configured to physically operate at least one of a polarizing plate rotation module, an analyzer insertion module, a stage rotation module, a focus adjustment module, and a magnification change module of the polarizing microscope device; and a central hub system that integrally manages polarizing microscope operation learning data and control algorithms corresponding to a plurality of polarizing microscope models and provides polarizing microscope operation learning data and control algorithms corresponding to the model of the polarizing microscope device to the human robot, wherein the central hub system comprises a data storage unit that stores polarizing microscope operation learning data and control algorithms for a polarizing microscope operation procedure performed in the mineral identification process of an experienced researcher, and polarizing microscope operation learning data corresponding to the model of the polarizing microscope device It includes a transmission unit that transmits a control algorithm to the human robot, and the human robot is configured to recognize the position or shape of the operation part of the polarizing microscope device and automatically perform operations related to the mineral identification process for the polarizing microscope device based on the polarizing microscope operation learning data and control algorithm received from the central hub system.

[0019] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the human robot is characterized by including a robot vision module for photographing the operating part of the polarizing microscope device and an operating part recognition module that recognizes the position of at least one of the polarizer rotation module, analyzer insertion module, stage rotation module, focus adjustment module, and magnification change module based on the operating part image acquired by the robot vision module.

[0020] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the human robot includes an end effector for gripping or rotating the operating part of the polarizing microscope device, and the end effector is configured to be replaceable or have a variable gripping width corresponding to at least one of a knob-type operating part, a lever-type operating part, a slide-type operating part, and a button-type operating part.

[0021] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the polarizing microscope operation learning data is characterized by including at least one of the polarizer rotation angle, whether an analyzer is inserted, stage rotation angle, focus adjustment amount, magnification change condition, and image capture time performed by an experienced researcher during the mineral identification process.

[0022] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the central hub system further comprises a model information storage unit that stores microscope model information including manufacturer information, model information, operation unit placement information, and operation unit driving method information of the polarizing microscope device.

[0023] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the central hub system further includes an operation mapping generation unit that generates operation mapping data including a reference position, operation direction, operation amount, and tolerance for each operation part of the polarizing microscope device based on the microscope model information, and the transmission unit transmits the operation mapping data, the polarizing microscope operation learning data, and a control algorithm together to the human robot, and the human robot performs operations related to the mineral identification process after correcting the difference in the position of the operation part according to the model difference of the polarizing microscope device based on the operation mapping data.

[0024] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the human robot includes a force sensing module that detects at least one of contact force, rotational resistance, movement resistance, and torque generated during the process of operating the operating part of the polarizing microscope device, and is characterized by adjusting the gripping force or operating speed for the operating part based on the detection result of the force sensing module.

[0025] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the system further includes an artificial intelligence analysis unit that analyzes a mineral image captured by the polarizing microscope device to calculate a mineral identification reliability, and the artificial intelligence analysis unit generates additional observation conditions when the mineral identification reliability is less than a reference reliability, and the central hub system transmits additional polarizing microscope operation learning data and a control algorithm corresponding to the additional observation conditions to the human robot, and the human robot performs additional operations on the polarizing microscope device based on the additional polarizing microscope operation learning data and control algorithm, and then acquires an additional mineral image.

[0026] In addition, in the physical AI-based human-robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the additional observation condition is characterized by including at least one of a stage rotation angle condition for confirming extinction characteristics, a polarizer rotation condition for confirming interference colors, a focus adjustment condition for confirming crystal boundaries, and a magnification change condition for confirming mineral grain magnification.

[0027] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the central hub system is characterized by further including a data collection unit that collects polarizing microscope operation history and captured images performed during the mineral identification process from a plurality of research institutions or analysis equipment.

[0028] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the central hub system further comprises an algorithm update unit that updates the polarizing microscope operation learning data and control algorithm using the polarizing microscope operation history and captured images collected by the data collection unit.

[0029] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the algorithm update unit extracts an operation sequence, operation angle, operation amount, and shooting time that improve mineral identification accuracy from the polarizing microscope operation history as a valid operation pattern based on whether the mineral identification result for the captured image is correct or reliable, and updates the valid operation pattern as a polarizing microscope operation scenario for each mineral type.

[0030] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the central hub system stores a mineral-specific operation scenario library including a polarizing microscope operation sequence required for each mineral type, and the mineral-specific operation scenario library includes at least two steps among an open Nicol observation step, an orthogonal Nicol observation step, a stage rotation observation step, a light extinction angle verification step, and an interference color verification step, and the transmission unit transmits a mineral-specific operation scenario corresponding to a candidate mineral group of a mineral sample to be analyzed to the human robot.

[0031] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the human robot performs a calibration operation to identify a reference position of the polarizing microscope device before performing an operation on the polarizing microscope device, and establishes a relative coordinate system between the polarizing microscope device and the human robot based on the result of the calibration operation.

[0032] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the human robot calculates an approach path for each operating part of the polarizing microscope device based on the relative coordinate system, and the approach path is configured to avoid interference with at least one of the eyepiece, objective lens part, sample stage, and imaging module of the polarizing microscope device.

[0033] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the central hub system further comprises an algorithm selection unit for selecting a version of polarizing microscope operation learning data and a control algorithm to be provided to the human robot based on identification information of the human robot, model information of the polarizing microscope device, and a mineral identification task type.

[0034] In addition, in the physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to the present invention, the central hub system is implemented as a cloud server, and the transmission unit provides polarizing microscope operation learning data and control algorithms to each of a plurality of human robots via a communication network, and each of the plurality of human robots performs operations related to the mineral identification process based on polarizing microscope operation learning data and control algorithms corresponding to models of different polarizing microscope devices.

[0035] Specific details of other embodiments are included in "Specific details for implementing the invention" and the attached "drawings".

[0036] The advantages and / or features of the present invention and the methods for achieving them will become clear by referring to the various embodiments described below in detail together with the accompanying drawings.

[0037] However, it should be understood that the present invention is not limited to the configurations of each embodiment disclosed below, but may be implemented in various different forms, and that each embodiment disclosed in this specification is provided merely to make the disclosure of the present invention complete and to fully inform those skilled in the art of the scope of the present invention, and that the present invention is defined only by the scope of each claim of the claims. Effects of the invention

[0038] The present invention can achieve the following effects through the combination and usage relationships of the embodiments of the present invention described above and the configuration described below.

[0039] According to the present invention, by automatically performing operations related to the mineral identification process, such as rotating the polarizer of a polarizing microscope, inserting and removing an analyzer, rotating the stage, adjusting the focus, and changing the magnification, the reliance on manual operation by the researcher can be reduced, and the time and workload required for repetitive microscope operations can be reduced.

[0040] In addition, according to the present invention, observation conditions suitable for mineral identification can be stably formed by utilizing polarizing microscope operation learning data and control algorithms built based on the mineral identification procedure of an experienced researcher, thereby enabling the acquisition of high-quality images suitable for AI-based mineral analysis and improving the accuracy and reproducibility of mineral identification.

[0041] Furthermore, according to the present invention, by integrally managing and providing operation learning data and control algorithms corresponding to polarizing microscopes of different manufacturers or models in a central hub system, it is possible to implement an automated mineral identification platform that is universally applicable in various research environments, and there is an effect of improving system performance by continuously learning and expanding mineral identification data accumulated from multiple research institutions. Brief explanation of the drawing

[0042] FIG. 1 is a schematic diagram showing a mineral identification support system according to one embodiment of the present invention. Figure 2 is a detailed configuration diagram showing the detailed configuration of the human robot of Figure 1. FIG. 3 is a schematic diagram showing a mineral identification support system according to another embodiment of the present invention. FIG. 4 is a schematic diagram showing a mineral identification support system according to another embodiment of the present invention. And, FIG. 5 is a flowchart illustrating a mineral identification support method according to one embodiment of the present invention. Specific details for implementing the invention

[0043] Before describing the present invention in detail, it should be understood that the terms and words used in this specification should not be interpreted as being limited to their ordinary or dictionary meanings, and that the inventor of the present invention may appropriately define and use the concepts of various terms to best describe their invention, and furthermore, that these terms and words should be interpreted in a meaning and concept consistent with the technical spirit of the present invention.

[0044] In other words, it should be understood that the terms used in this specification are used merely to describe preferred embodiments of the present invention and are not intended to specifically limit the content of the present invention, and that these terms are defined in consideration of various possibilities of the present invention.

[0045] In addition, it should be noted that in this specification, singular expressions may include plural expressions unless the context clearly indicates a different meaning, and that even if they are expressed in a similarly plural form, they may include a singular meaning.

[0046] Throughout this specification, where it is stated that a component "includes" another component, unless specifically stated otherwise, this may mean that it does not exclude any other component but may include any other component.

[0047] Furthermore, it should be noted that where it is stated that a component "exists inside or is installed in connection with" another component, this component may be installed in direct connection or contact with the other component, or it may be installed at a certain distance apart; in the case where it is installed at a certain distance apart, a third component or means for fixing or connecting the component to the other component may exist, and a description of this third component or means may be omitted.

[0048] On the other hand, where it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that a third component or means does not exist.

[0049] Likewise, other expressions describing the relationship between each component, such as “between” and “right between”, or “adjacent to” and “directly adjacent to”, should be interpreted as having the same intent.

[0050] In addition, it should be understood that in this specification, terms such as "one side," "other side," "one side," "other side," "first," and "second," if used, are intended to clearly distinguish one component from another component, and that the meaning of the component is not restricted by such terms.

[0051] In addition, position-related terms such as "up," "down," "left," and "right" used in this specification should be understood as indicating the relative position of the corresponding component in the drawing, and unless an absolute position is specified, these position-related terms should not be understood as referring to an absolute position.

[0052] Furthermore, in specifying the reference numerals for each component of each drawing in this specification, the same component has the same reference numeral even if it is shown in different drawings; that is, the same reference numeral throughout the specification indicates the same component.

[0053] In the drawings attached to this specification, the size, position, connection relationships, etc., of each component constituting the present invention may be described in a partially exaggerated, reduced, or omitted manner for the convenience of explanation or to sufficiently clearly convey the concept of the present invention, and therefore, the proportions or scale may not be strictly accurate.

[0054] In addition, in the following description of the present invention, detailed descriptions of components that are deemed to unnecessarily obscure the essence of the present invention, such as known technologies including prior art, may be omitted.

[0056] Preferred embodiments of the present invention will be described in detail below with reference to the drawings.

[0057] FIG. 1 is a schematic diagram showing a mineral identification support system according to one embodiment of the present invention.

[0058] As illustrated in FIG. 1, a mineral identification support system (100) according to one embodiment of the present invention is a physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system for identifying the type of mineral included in a mineral sample, and may include a polarizing microscope device (110) for observing a mineral sample, a human robot (120) configured to physically operate the polarizing microscope device (110), and a central hub system (130) that integrally manages polarizing microscope operation learning data and control algorithms corresponding to a plurality of polarizing microscope models and provides polarizing microscope operation learning data and control algorithms corresponding to the model of the polarizing microscope device (110) to the human robot (120).

[0059] The polarizing microscope device (110) is a device for observing mineral samples, and may observe mineral thin sections, rock thin sections, mineral samples in a granular state, or mineral samples in a powdered state. The polarizing microscope device (110) may be configured to observe at least one of the interference color, birefringence characteristics, extinction characteristics, crystal form, grain boundaries, twinning structure, or optical anisotropy of the mineral sample, and the results of such mineralogical observations may be used as basic data for identifying the type of mineral contained in the mineral sample.

[0060] A polarizing microscope device (110) may include at least one of a polarizer rotation module, an analyzer insertion module, a stage rotation module, a focus adjustment module, and a magnification change module to perform microscope operations required in the mineral identification process. The polarizer rotation module may be configured to change the polarization conditions incident on the mineral sample, and the analyzer insertion module may be configured to switch between an open Nicol observation state and an orthogonal Nicol observation state. The stage rotation module may be configured to change the observation direction or crystal orientation of the mineral sample, the focus adjustment module may be configured to adjust the focus position for the surface, grain boundary, or internal region of the mineral sample, and the magnification change module may be configured to selectively magnify and observe the entire structure or a specific grain region of the mineral sample.

[0061] The human robot (120) may be configured to physically operate at least one of the polarizer rotation module, analyzer insertion module, stage rotation module, focus adjustment module, and magnification change module of the polarizing microscope device (110). The human robot (120) may include a robot arm, robot hand, gripper, end effector, or multi-joint drive structure capable of mimicking human arm or hand movements, and may approach the operation part of the polarizing microscope device (110) to perform rotation, grasping, pressing, pulling, insertion, removal, or movement actions. Accordingly, the human robot (120) can automatically perform operations related to the mineral identification process, such as polarizer rotation, analyzer insertion and removal, stage rotation, focus adjustment, and magnification change, which were previously performed manually by a researcher.

[0062] The central hub system (130) may be configured to integrally manage polarization microscope operation learning data and control algorithms corresponding to multiple polarization microscope models. The central hub system (130) may be implemented in the form of a local server installed within a single analysis room, or in the form of a cloud server connected via a communication network to multiple research institutions or multiple laboratories. The central hub system (130) may manage information corresponding to the manufacturer, model, arrangement of control parts, type of control parts, driving method for each control part, and allowable operation range for each control part of the polarization microscope device (110), and based on this, may provide polarization microscope operation learning data and control algorithms suitable for the model of the polarization microscope device (110) to the human robot (120).

[0063] The central hub system (130) may include a data storage unit (131) and a transmission unit (132). The data storage unit (131) may store polarization microscope operation learning data and control algorithms for the polarization microscope operation procedures performed during the mineral identification process by an experienced researcher. For example, the data storage unit (131) may store at least one of the following in correspondence with each other: the polarizer rotation angle, whether an analyzer is inserted, the stage rotation angle, the amount of focus adjustment, the magnification change condition, the time of image capture, the order of observation, the mineral candidate group, and the final identification result performed by the experienced researcher when identifying a specific mineral sample.

[0064] Additionally, the data storage unit (131) can store operation mapping data corresponding to a plurality of polarizing microscope models. The operation mapping data may include position information, shape information, operation direction information, operation amount information, access path information, and tolerance information for a polarizing plate rotation module, an analyzer insertion module, a stage rotation module, a focus adjustment module, and a magnification change module for each model of the polarizing microscope device (110). Accordingly, even if the manufacturers or models of the polarizing microscope devices (110) are different, the human robot (120) can recognize the operation part of the corresponding polarizing microscope device (110) and perform appropriate operations based on the operation mapping data provided from the central hub system (130).

[0065] The transmission unit (132) may be configured to transmit polarizing microscope operation learning data and control algorithms corresponding to the model of the polarizing microscope device (110) to the human robot (120). The transmission unit (132) may receive model information of the polarizing microscope device (110), mineral sample information, work type information, or information on candidate mineral groups for identification from the human robot (120), and based on the received information, may select data suitable for the corresponding work from among the polarizing microscope operation learning data and control algorithms stored in the data storage unit (131) and transmit it to the human robot (120). Additionally, the transmission unit (132) may be configured to sequentially transmit the necessary polarizing microscope operation learning data and control algorithms according to the progress stage of the mineral identification process.

[0066] The human robot (120) can recognize the position or shape of the operation part of the polarizing microscope device (110) based on polarizing microscope operation learning data and control algorithms received from the central hub system (130). For example, the human robot (120) can identify the position or shape of the polarizer rotation module, analyzer insertion module, stage rotation module, focus adjustment module, and magnification change module of the polarizing microscope device (110) using at least one of camera-based recognition, sensor-based recognition, distance measurement, shape recognition, or reference point calibration. Subsequently, the human robot (120) can establish a relative coordinate system between the robot coordinate system and the device coordinate system of the polarizing microscope device (110) based on the identified position of the operation part, and can automatically perform operations related to the mineral identification process based on the established relative coordinate system.

[0067] In one embodiment, the human robot (120) can change the operation operation by receiving polarizing microscope operation learning data and control algorithms corresponding to the model from the central hub system (130) even when the model of the polarizing microscope device (110) is changed. For example, even when the focus adjustment module is arranged in the form of a side knob in the polarizing microscope device (110) of the first model and the focus adjustment module is arranged in the form of a front dial in the polarizing microscope device (110) of the second model, the human robot (120) can perform focus adjustment by changing the approach position, gripping direction, rotation direction, and amount of rotation according to the operation mapping data corresponding to each model.

[0068] In another embodiment, the human robot (120) can automatically operate the polarizing microscope device (110) according to a standardized observation scenario during the mineral identification process. For example, the human robot (120) can sequentially perform at least two steps among an open Nicol observation step, an orthogonal Nicol observation step, a polarizer rotation step, an analyzer insertion or removal step, a stage rotation step, a focus adjustment step, a magnification change step, and an image capture step. In this case, the image obtained by the polarizing microscope device (110) may be an image that reflects the observation conditions required for mineral identification, and the accuracy and reproducibility of mineral identification may be improved during subsequent artificial intelligence analysis or researcher review.

[0069] In another embodiment, the data storage unit (131) may store operation scenarios for each mineral type. The operation scenario for each mineral type may include at least one of polarization conditions, stage rotation conditions, extinction angle verification conditions, interference color verification conditions, focus adjustment conditions, and magnification change conditions required to distinguish a specific mineral or candidate mineral group. The transmission unit (132) may transmit the operation scenario for each mineral type corresponding to the candidate mineral group of the mineral sample to be analyzed to the human robot (120), and the human robot (120) may operate the polarizing microscope device (110) according to the received operation scenario for each mineral type to acquire an image necessary for the identification of the corresponding mineral.

[0070] In another embodiment, the mineral identification support system (100) may be configured to provide a mineral image acquired from a polarizing microscope device (110) to an artificial intelligence analysis module. The artificial intelligence analysis module may calculate a mineral identification result or a mineral identification reliability based on the mineral image, and may generate additional observation conditions if the mineral identification reliability is below a reference reliability. In this case, the central hub system (130) may transmit polarizing microscope operation learning data and control algorithms corresponding to the additional observation conditions back to the human robot (120) through the transmission unit (132), and the human robot (120) may enable the polarizing microscope device (110) to acquire additional mineral images by performing additional polarizer rotation, additional analyzer insertion or removal, additional stage rotation, additional focus adjustment, or additional magnification change.

[0071] In another embodiment, the central hub system (130) can collect polarizing microscope operation history, captured images, identification results, and analysis reliability information from multiple research institutions or multiple analysis equipment. The central hub system (130) can accumulate and store the collected information in a data storage unit (131) and update polarizing microscope operation learning data and control algorithms based on the accumulated information. The updated polarizing microscope operation learning data and control algorithms can be provided to the human robot (120) through the delivery unit (132), and accordingly, the mineral identification support system (100) can be expanded into a platform structure in which performance is continuously improved based on data accumulated from multiple polarizing microscope devices (110), multiple human robots (120), and multiple research institutions.

[0072] In this way, the mineral identification support system (100) can automate the physical operation of the polarizing microscope device (110) required in the mineral identification process through the interlocking structure of the polarizing microscope device (110), the human robot (120), and the central hub system (130). In particular, the data storage unit (131) of the central hub system (130) stores polarizing microscope operation learning data and control algorithms regarding the polarizing microscope operation procedure of a skilled researcher, and the transmission unit (132) transmits the polarizing microscope operation learning data and control algorithms corresponding to the model of the polarizing microscope device (110) to the human robot (120), and the human robot (120) can automatically perform operations related to the mineral identification process after recognizing the position or shape of the operation part of the polarizing microscope device (110) based on the received polarizing microscope operation learning data and control algorithms. Accordingly, the mineral identification support system (100) can reduce manual operation deviations by researchers and provide a physical AI-based mineral identification automation structure capable of responding to various models of polarizing microscope devices (110).

[0073] Figure 2 is a detailed configuration diagram showing the detailed configuration of the human robot of Figure 1.

[0074] As illustrated in FIG. 2, the human robot (120) of FIG. 1 may include a robot vision module (121), a control part recognition module (122), an end effector (123), and a force detection module (124) as a configuration to recognize a control part of a polarizing microscope device (110), perform physical operation on the recognized control part, and detect a physical reaction force generated during the operation process to ensure operation stability.

[0075] A robot vision module (121) may be configured to photograph the operating part of a polarizing microscope device (110). The robot vision module (121) may be positioned adjacent to the head, arm, wrist, or end effector (123) of a human robot (120) and may photograph the appearance of the polarizing microscope device (110), the location of the operating part, the shape of the operating part, and interference structures around the operating part. Here, the operating part may be a physical operating area corresponding to at least one of the polarizer rotation module, analyzer insertion module, stage rotation module, focus adjustment module, and magnification change module of the polarizing microscope device (110). For example, the robot vision module (121) may photograph a rotating ring or dial corresponding to the polarizer rotation module, a slider or lever corresponding to the analyzer insertion module, a rotating knob corresponding to the stage rotation module, a coarse or fine adjustment screw corresponding to the focus adjustment module, and an objective lens switching part or magnification selection part corresponding to the magnification change module.

[0076] The control unit recognition module (122) may be configured to recognize the position of at least one of the polarizer rotation module, analyzer insertion module, stage rotation module, focus adjustment module, and magnification change module of the polarizing microscope device (110) based on the control unit image acquired by the robot vision module (121). The control unit recognition module (122) can analyze the outer shape, center position, rotation axis direction, movement direction, degree of protrusion, color contrast, reference marker, or character mark of the control unit in the control unit image, and can calculate the control coordinates that the human robot (120) must approach based on the results of such analysis. Additionally, the control unit recognition module (122) can correct the installation position error of the polarizing microscope device (110), the difference in control unit placement by model, or the difference in control unit shape by comparing the model-specific control mapping data of the polarizing microscope device (110) provided by the central hub system (130) with the actual control unit image acquired by the robot vision module (121).

[0077] The control unit recognition module (122) may be configured not only to simply detect the position of the control unit of the polarizing microscope device (110), but also to distinguish the type of operation of the control unit. For example, the control unit recognition module (122) can determine whether a specific control unit is a knob-type control unit requiring rotational operation, a slide-type control unit requiring linear movement, a lever-type control unit requiring insertion or removal operation, or a button-type control unit requiring a pressing operation. Accordingly, the human robot (120) can determine the approach direction, gripping position, gripping force, rotation direction, amount of rotation, distance traveled, or operation speed of the end effector (123) based on the recognition result of the control unit recognition module (122).

[0078] The end effector (123) may be configured to grasp or rotate the control part of the polarizing microscope device (110). The end effector (123) may be placed at the end of the human robot (120) and may physically operate the control part corresponding to the polarizing plate rotation module, analyzer insertion module, stage rotation module, focus adjustment module, and magnification change module of the polarizing microscope device (110) by directly contacting it. For example, the end effector (123) may rotate the knob-type control part by a predetermined angle while grasping it, insert or remove the analyzer by pushing or pulling the lever-type control part, move the slide-type control part in a straight direction, and perform magnification change or shooting condition setting by pressing the button-type control part.

[0079] The end effector (123) may be configured to be replaceable or have a variable gripping width corresponding to at least one of a knob-type control part, a lever-type control part, a slide-type control part, and a button-type control part. For example, the end effector (123) may selectively be equipped with at least one of an arc-shaped gripping part for stably gripping a rotary knob, a hook-type contact part for pushing or pulling a lever, a flat contact part for moving a slider in a straight line, and an elastic pressing part for pressing a button. Additionally, the end effector (123) may adjust the gripping width according to the size or shape of the control part, so that the same human robot (120) can perform physical operations on polarizing microscope devices (110) of different manufacturers or models.

[0080] The force sensing module (124) may be configured to detect at least one of contact force, rotational resistance, movement resistance, and torque generated during the process in which the human robot (120) operates the operating part of the polarizing microscope device (110). The force sensing module (124) may be placed on the end effector (123), the wrist part, joint part, or drive part of the human robot (120), and may detect the contact force at the moment the human robot (120) contacts the operating part, the rotational resistance during the process of rotating the operating part, the movement resistance during the process of moving the slide-type operating part, and the torque during the process of rotating the knob-type operating part.

[0081] The human robot (120) can adjust the gripping force or operating speed of the operating part based on the detection result of the force detection module (124). For example, the human robot (120) can increase the gripping force or decrease the rotation speed when the rotational resistance is greater than a reference value during the process of rotating the knob-type operating part, and can temporarily stop the operation to prevent damage to the operating part when the rotational resistance is abnormally large. In addition, the human robot (120) can stop the pressing operation when the contact force exceeds a reference value during the process of pressing the button-type operating part, and can reset the movement path when the movement resistance exceeds a reference range during the process of moving the slide-type operating part. Accordingly, the human robot (120) can stably operate the operating part of the polarizing microscope device (110) while preventing damage to the operating part, excessive force application, or erroneous operation.

[0082] In one embodiment, the human robot (120) can acquire an image of the operating part of the polarizing microscope device (110) by the robot vision module (121), and after recognizing the position of the target module for operation among the polarizer rotation module, analyzer insertion module, stage rotation module, focus adjustment module, and magnification change module by the operating part recognition module (122), the end effector (123) can be brought close to the corresponding operating part. Subsequently, the end effector (123) can perform gripping, rotation, pressing, pulling, or moving actions depending on the type of operating part, and the force detection module (124) can be used to correct the operation action of the human robot (120) by detecting contact force, rotational resistance, movement resistance, or torque generated during the operation process.

[0083] In another embodiment, the human robot (120) can sequentially select target modules for operation based on polarizing microscope operation learning data and control algorithms provided by the central hub system (130). For example, if open Nicol observation is required during the mineral identification process, the human robot (120) can operate the analyzer insertion module to the removal position; if orthogonal Nicol observation is required, the analyzer insertion module can be operated to the insertion position; and if confirmation of extinction characteristics is required, the stage rotation module can be rotated at predetermined angle intervals. Additionally, if confirmation of interference color is required, the human robot (120) can operate the polarizer rotation module; and if confirmation of particle boundaries or crystal shapes is required, the focus adjustment module and magnification change module can be operated.

[0084] In another embodiment, the robot vision module (121) and the control unit recognition module (122) may be configured to re-recognize the position or shape of the control unit of the polarizing microscope device (110) even when the model of the polarizing microscope device (110) is changed. In this case, the human robot (120) can update the target position for operation based on the actual captured image of the control unit and model-specific operation mapping data provided from the central hub system (130), rather than relying only on the fixed control coordinates set on the specific polarizing microscope device (110). Accordingly, the human robot (120) can flexibly respond to polarizing microscope devices (110) of different manufacturers, models, or installation environments.

[0085] In another embodiment, the end effector (123) may include a plurality of interchangeable end tips, and the human robot (120) may select a suitable end tip according to the type of operation recognized by the operation recognition module (122). For example, if the focus adjustment module is configured as a knob-type operation requiring fine rotation, the end effector (123) may use a rotary operation end tip having an anti-slip gripping surface, and if the analyzer insertion module is configured as a slide-type operation, the end effector (123) may use an end tip having a contact surface for linear movement. With this configuration, the human robot (120) can correspond to the shape of the operation of various polarizing microscope devices (110) as well as a single polarizing microscope device (110).

[0086] In another embodiment, the force sensing module (124) can be used to determine the state of the operating part of the polarizing microscope device (110). For example, if the rotational resistance detected while the human robot (120) is rotating the focus adjustment module is abnormally low, it can be determined that the operating part has failed to grip or is slipping, and if the rotational resistance is abnormally high, it can be determined that the operating limit position has been reached or that there is mechanical interference. In this case, the human robot (120) can correct the gripping position of the end effector (123), reduce the operating speed, stop the operation, and transmit the error state to the central hub system (130).

[0087] In another embodiment, the human robot (120) can perform operations related to the mineral identification process in a closed-loop manner through the mutual interoperability of the robot vision module (121), the operation part recognition module (122), the end effector (123), and the force sensing module (124). That is, the human robot (120) can check the position of the operation part before operation through the robot vision module (121), physically operate the operation part through the end effector (123), check the physical state during operation through the force sensing module (124), and after the operation is completed, check the final position of the operation part or the state of the polarizing microscope device (110) again through the robot vision module (121). Accordingly, the human robot (120) can perform physical AI-based operations that combine visual recognition and force sensing, rather than simple open-loop robot movements.

[0088] FIG. 3 is a schematic diagram showing a mineral identification support system according to another embodiment of the present invention.

[0089] As illustrated in FIG. 3, a mineral identification support system (100) according to another embodiment of the present invention is a physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system for identifying the type of mineral contained in a mineral sample, and may include a polarizing microscope device (110) for observing a mineral sample, a human robot (120) configured to physically operate the polarizing microscope device (110), and a central hub system (130) that integrally manages polarizing microscope operation learning data and control algorithms reflecting the model-specific characteristics of the polarizing microscope device (110) and mineral identification work conditions. The embodiment of FIG. 3 is an example in which the internal configuration of the central hub system (130) is shown in a more expanded form compared to the embodiment of FIG. 1, and the central hub system (130) may not only perform the function of simply storing and transmitting data, but also perform the function of managing the characteristics of the polarizing microscope device (110) by model, generating operation mapping data, collecting data from multiple research institutions or analysis equipment, updating algorithms using the collected data, and selecting algorithms by human robot (120) and by work type.

[0090] The polarizing microscope device (110) may be a device for observing mineral thin sections, rock thin sections, mineral particles, or mineral samples in powder form, and may be configured to observe at least one of the interference color, birefringence properties, extinction properties, crystal form, grain boundaries, and optical anisotropy of the mineral contained in the mineral sample. The polarizing microscope device (110) may include at least one of a polarizer rotation module, an analyzer insertion module, a stage rotation module, a focus adjustment module, and a magnification change module, and these modules may be targets for physical manipulation to form observation conditions necessary for the mineral identification process. For example, the polarizer rotation module may be used to change polarization conditions, the analyzer insertion module may be used to switch between open Nicol states and orthogonal Nicol states, and the stage rotation module may be used for observing extinction properties or crystal orientation. Additionally, the focus adjustment module may be used to focus on a specific depth or grain boundary of the mineral sample, and the magnification change module may be used for observing the entire texture of the mineral sample or for magnified observation of a specific crystal.

[0091] The human robot (120) may be configured to recognize the position or shape of the operation part of the polarizing microscope device (110) based on polarizing microscope operation learning data and control algorithms provided by the central hub system (130), and to automatically perform operations related to the mineral identification process for the polarizing microscope device (110). In particular, in the embodiment of FIG. 3, the human robot (120) may be configured to perform a calibration operation to identify the reference position of the polarizing microscope device (110) before performing the operation. For example, the human robot (120) may identify the reference position of the polarizing microscope device (110) by photographing or sensing a specific reference point of the polarizing microscope device (110), such as the center point of the sample stage, a reference corner of the outer edge of the main body, a reference position of the operation panel, or an optical axis alignment reference part. And the human robot (120) can establish a relative coordinate system between the polarizing microscope device (110) and the human robot (120) based on the result of the calibration operation. Accordingly, the human robot (120) can perform more accurate operations based on a relative coordinate system that reflects the actual installation state, rather than simply using pre-fixed absolute coordinates.

[0092] Additionally, the human robot (120) can calculate an approach path for each operating part of the polarizing microscope device (110) based on the relative coordinate system. At this time, the approach path can be set to avoid interference with at least one of the eyepiece, objective lens, sample stage, and imaging module of the polarizing microscope device (110). For example, when the human robot (120) approaches to operate the stage rotation module, a lateral approach path may be selected so as not to collide with the objective lens, and when operating the focus adjustment module, a detour path may be set to avoid interference with the imaging module or the eyepiece. By setting such an approach path, the human robot (120) can stably perform operations on polarizing microscope devices (110) of various shapes.

[0093] The central hub system (130) may include a data storage unit (131), a transmission unit (132), a model information storage unit (133), an operation mapping generation unit (134), a data collection unit (135), an algorithm update unit (136), and an algorithm selection unit (137). The data storage unit (131) may store polarization microscope operation learning data and control algorithms for the polarization microscope operation procedure performed by an experienced researcher during the mineral identification process. Here, the polarization microscope operation learning data may include at least one of the polarizer rotation angle, whether an analyzer is inserted, the stage rotation angle, the amount of focus adjustment, the magnification change condition, and the time of image capture performed by the experienced researcher during the mineral identification process. For example, information such as when an analyzer was inserted and within what angle range the stage was rotated to identify a specific mineral, at what magnification the image was captured, and how much the polarizer rotation angle was set to to verify specific optical properties may be stored in the data storage unit (131). This data is not merely a set of simple operation commands, but can function as mineral identification procedure data that reflects the practical judgment of skilled researchers.

[0094] The model information storage unit (133) can store microscope model information including manufacturer information, model information, operation unit placement information, and operation unit driving method information of the polarizing microscope device (110). For example, in the first model of the first manufacturer, the focus adjustment module may be arranged in a double knob structure on the right side, and in the second model of the second manufacturer, the focus adjustment module may be arranged in a front dial structure, and the analyzer insertion module may also be implemented in different driving methods such as a slide type, a lever type, or a button type. The model information storage unit (133) can support the human robot (120) to correspond to different polarizing microscope devices (110) by creating a database of such model-specific differences.

[0095] The operation mapping generation unit (134) can generate operation mapping data including reference position, operation direction, amount of operation, and tolerance for each operation part of the polarizing microscope device (110) based on microscope model information stored in the model information storage unit (133). For example, the operation mapping generation unit (134) can generate operation mapping data for a specific model's stage rotation module including reference position coordinates, operation direction information indicating that clockwise rotation is the increasing direction, actual amount of operation information corresponding to a rotation command in 10-degree increments, and an allowable position error or rotation error range. Additionally, the operation mapping generation unit (134) can generate individual mapping data for each operation part by distinguishing between the coarse and fine adjustment screws of the focus adjustment module, and in the case of the analyzer insertion module, it can generate mapping data defining the insertion completion position and the removal completion position. The transmission unit (132) can transmit the operation mapping data, the polarizing microscope operation learning data, and the control algorithm together to the human robot (120), and the human robot (120) can perform operations related to the mineral identification process after correcting the difference in the position of the operation unit according to the model difference of the polarizing microscope device (110) based on the operation mapping data. Accordingly, even for the same mineral identification task, it is possible to perform operations that reflect different positions of the operation unit and driving methods depending on the model of the polarizing microscope device (110).

[0096] The transmission unit (132) can perform the function of transmitting polarizing microscope operation learning data and control algorithms stored in the data storage unit (131), and operation mapping data generated by the operation mapping generation unit (134), to the human robot (120). The transmission unit (132) may transmit the entire operation scenario collectively at a single point in time, or it may transmit only the necessary step data sequentially according to the progress of the mineral identification process. For example, it may be implemented in such a way that data corresponding to the open Nicol observation stage is transmitted first, and then data corresponding to the orthogonal Nicol observation stage or the extinction angle verification stage is subsequently transmitted.

[0097] The data collection unit (135) can collect the operation history and captured images of the polarizing microscope performed during the mineral identification process from multiple research institutions or analysis equipment. For example, the operation history and captured images performed by the first polarizing microscope device (110) and the first human robot (120) of the first research institution, and the operation history and captured images performed by the second polarizing microscope device (110) and the second human robot (120) of the second research institution can be collected by the data collection unit (135). Here, the operation history may include the polarizing plate rotation sequence, the insertion and removal history of the analyzer, the stage rotation angle history, the focus adjustment amount history, the magnification change sequence, the time of shooting, and the time of execution for each step, and the captured images may be polarizing microscope images obtained for each operation step.

[0098] The algorithm update unit (136) can update the polarizing microscope operation learning data and control algorithm using the polarizing microscope operation history and captured images collected by the data collection unit (135). In particular, the algorithm update unit (136) can extract the operation sequence, operation angle, amount of operation, and time of shooting that improved the mineral identification accuracy among the polarizing microscope operation history as valid operation patterns based on whether the mineral identification result for the captured image is correct or reliable. For example, if an image captured under conditions of open Nicol observation followed by orthogonal Nicol observation and a specific focus adjustment amount, after performing open Nicol observation for a specific mineral group, shows high identification accuracy, the algorithm update unit (136) can extract the corresponding operation sequence, operation angle, amount of operation, and time of shooting as valid operation patterns. The algorithm update unit (136) can update the valid operation patterns as polarizing microscope operation scenarios for each mineral type. Accordingly, the scenario stored in the data storage unit (131) does not remain at a fixed initial rule, but can be continuously improved by reflecting actual operation results.

[0099] The central hub system (130) can store a mineral-specific operation scenario library containing the required polarizing microscope operation sequence for each mineral type. The mineral-specific operation scenario library may include at least two of the following steps: an open Nicol observation step, an orthogonal Nicol observation step, a stage rotation observation step, an extinction angle verification step, and an interference color verification step. For example, a mineral-specific operation scenario for quartz may include an open Nicol observation step, an orthogonal Nicol observation step, and an extinction angle verification step, and a mineral-specific operation scenario for mica may include an open Nicol observation step, an orthogonal Nicol observation step, an interference color verification step, and a stage rotation observation step. The delivery unit (132) can transmit a mineral-specific operation scenario corresponding to a candidate mineral group of the mineral sample to be analyzed to a human robot (120), and accordingly, the human robot (120) can automatically perform a mineral identification procedure optimized for the candidate mineral group.

[0100] The algorithm selection unit (137) can select a version of polarizing microscope operation learning data and control algorithm to be provided to the human robot (120) based on the identification information of the human robot (120), the model information of the polarizing microscope device (110), and the type of mineral identification task. For example, even if the same type of mineral is targeted, the end effector configuration of the first human robot (120), the operating range of the second human robot (120), the arrangement of the operation part of the first model polarizing microscope device (110), the driving method of the second model polarizing microscope device (110), and the type of task, such as whether it is thin section sample analysis or powder sample analysis, may differ from each other. Therefore, the algorithm selection unit (137) can select an optimal algorithm version according to these conditions and provide it through the delivery unit (132). Accordingly, the system (100) may have an adaptive algorithm provision structure that reflects both hardware differences and work purpose differences.

[0101] In another embodiment of FIG. 3, the central hub system (130) may be implemented as a cloud server, and the delivery unit (132) may provide polarizing microscope operation learning data and control algorithms to each of the plurality of human robots (120) via a communication network. In this case, each of the plurality of human robots (120) may perform operations related to the mineral identification process based on polarizing microscope operation learning data and control algorithms corresponding to models of different polarizing microscope devices (110). For example, the human robot (120) of the first regional research institution may be operated by receiving an algorithm corresponding to the first model of the polarizing microscope device (110), and the human robot (120) of the second regional research institution may be operated by receiving an algorithm corresponding to the second model of the polarizing microscope device (110). Through such a cloud-based structure, the mineral identification support system (100) can be expanded beyond the level of a single laboratory into a platform as a service that encompasses multiple regions, multiple equipment, and multiple robots.

[0102] In another embodiment, the captured images and operation history collected by the data collection unit (135) are not merely used for central storage, but can also be utilized by the algorithm update unit (136) to generate new versions of polarizing microscope operation learning data and control algorithms. Additionally, the algorithm selection unit (137) may be configured to selectively distribute only verified versions to the human robot (120) after generating new versions. By doing so, experimental versions and operational versions can be managed separately, thereby further improving system stability in actual research sites.

[0103] In another embodiment, the operation mapping generation unit (134) may dynamically modify the operation mapping data by reflecting the calibration results of the human robot (120). For example, even if the polarizing microscope device (110) is of the same model, the actual position of the operation unit may differ slightly depending on the installation location, the height of the equipment stand, whether an additional shooting module is installed, etc. Therefore, the operation mapping generation unit (134) may fine-tune the operation mapping data by reflecting the calibration results transmitted from the human robot (120) or the history of successful operations. This dynamic correction structure may be advantageous for improving the success rate of operations.

[0104] FIG. 4 is a schematic diagram showing a mineral identification support system according to another embodiment of the present invention.

[0105] As illustrated in FIG. 4, a mineral identification support system (100) according to another embodiment of the present invention is a system for identifying the type of mineral contained in a mineral sample, and may include a polarizing microscope device (110) for observing a mineral sample, a human robot (120) configured to physically operate the polarizing microscope device (110), a central hub system (130) that provides polarizing microscope operation learning data and a control algorithm corresponding to a model of the polarizing microscope device (110) to the human robot (120), and an artificial intelligence analysis unit (140) that analyzes a mineral image captured by the polarizing microscope device (110) to calculate the reliability of mineral identification. This embodiment is a structure in which an artificial intelligence analysis unit (140) is additionally linked to the mineral identification support system (100) of FIG. 3, and may represent a configuration in which mineral images are acquired through a polarizing microscope device (110), mineral identification reliability is calculated through the artificial intelligence analysis unit (140), additional operation data is provided through a central hub system (130), and additional operations are performed through a human robot (120) in a closed-loop manner. The description of such detailed embodiments may be configured to be consistent with the intent of the examination guidelines, which require that the description of the invention be specifically described to support the claims and enable a person skilled in the art to easily implement it.

[0106] The polarizing microscope device (110) may be configured to photograph mineral samples in the form of mineral thin sections, rock thin sections, mineral particles, or powders, and may acquire a mineral image reflecting at least one of polarization conditions, analyzer insertion conditions, stage rotation conditions, focus conditions, and magnification conditions for the mineral sample. The mineral image captured by the polarizing microscope device (110) may be an image in an open Nicol state, an image in an orthogonal Nicol state, an image taken at a specific stage rotation angle, an image taken at a specific focus position, or an image taken under specific magnification conditions. At this time, the mineral image may be a single image, or a set of multiple images taken sequentially under different polarization conditions or different observation conditions for the same mineral sample.

[0107] The human robot (120) can physically operate at least one of the polarizing plate rotation module, analyzer insertion module, stage rotation module, focus adjustment module, and magnification change module of the polarizing microscope device (110) based on polarizing microscope operation learning data and control algorithms provided by the central hub system (130). The human robot (120) can automatically operate the polarizing microscope device (110) during the initial observation stage of the mineral identification process to obtain the first mineral image, and subsequently, if additional observation is required according to the analysis results of the artificial intelligence analysis unit (140), it can receive additional polarizing microscope operation learning data and control algorithms from the central hub system (130) and perform additional operations on the polarizing microscope device (110).

[0108] The central hub system (130) may include a data storage unit (131), a transmission unit (132), a model information storage unit (133), an operation mapping generation unit (134), a data collection unit (135), an algorithm update unit (136), and an algorithm selection unit (137). The data storage unit (131) may store polarizing microscope operation learning data and control algorithms performed during the mineral identification process by a skilled researcher, and the transmission unit (132) may transmit polarizing microscope operation learning data and control algorithms corresponding to the model of the polarizing microscope device (110) to the human robot (120). Additionally, the model information storage unit (133) may store manufacturer information, model information, operation unit placement information, and operation unit driving method information of the polarizing microscope device (110), and the operation mapping generation unit (134) may generate operation mapping data including the operation unit location, operation direction, operation amount, and tolerance for each model of the polarizing microscope device (110). The data collection unit (135) can collect polarizing microscope operation history and captured images performed during the mineral identification process, the algorithm update unit (136) can update polarizing microscope operation learning data and control algorithms using the collected data, and the algorithm selection unit (137) can select an algorithm version to be provided based on the identification information of the human robot (120), the model information of the polarizing microscope device (110), and the type of mineral identification work.

[0109] The artificial intelligence analysis unit (140) may be configured to analyze a mineral image captured by a polarizing microscope device (110) to calculate a mineral identification result and a mineral identification reliability. The artificial intelligence analysis unit (140) can extract image features corresponding to the color distribution, interference color pattern, extinction change, crystal boundary, grain shape, texture pattern, birefringence characteristics, or optical anisotropy of the mineral included in the mineral image, and can calculate a candidate group of minerals included in the mineral sample based on the extracted image features. Additionally, the artificial intelligence analysis unit (140) can calculate a probability value, a similarity value, or a reliability value for each of the calculated candidate groups, and can output the candidate mineral having the highest value among them as a provisional identification result.

[0110] The artificial intelligence analysis unit (140) can generate additional observation conditions when the mineral identification reliability is below the reference reliability. Here, the reference reliability may be a pre-set threshold value and may be set differently depending on the type of mineral, the purpose of analysis, the condition of the sample, the quality of the captured image, or the operational policy of the research institution. For example, if the mineral identification reliability is calculated to be below the reference reliability due to similar image features during the process of distinguishing between feldspar and quartz by the artificial intelligence analysis unit (140), the artificial intelligence analysis unit (140) may determine that additional orthogonal Nicol observation, additional stage rotation observation, or additional focus adjustment observation is required. In this case, the artificial intelligence analysis unit (140) can generate additional observation conditions and provide them to the central hub system (130).

[0111] The above additional observation conditions may include at least one of a stage rotation angle condition for verifying extinction characteristics, a polarizer rotation condition for verifying interference colors, a focus adjustment condition for verifying crystal boundaries, and a magnification change condition for verifying magnified mineral particles. For example, the artificial intelligence analysis unit (140) may generate a stage rotation angle condition as an additional observation condition when the extinction characteristics in the mineral image are unclear, and may generate a polarizer rotation condition as an additional observation condition when the interference colors are unclear. Additionally, the artificial intelligence analysis unit (140) may generate a focus adjustment condition as an additional observation condition when the crystal boundaries are unclear or the contours of the mineral particles are blurry, and may generate a magnification change condition as an additional observation condition when it is necessary to verify the fine crystal shape or twin structure of the mineral particles.

[0112] The central hub system (130) receives additional observation conditions generated by the artificial intelligence analysis unit (140) and can transmit additional polarizing microscope operation learning data and control algorithms corresponding to the additional observation conditions to the human robot (120). For example, if the additional observation condition is a stage rotation angle condition for verifying extinction characteristics, the central hub system (130) can select additional polarizing microscope operation learning data and control algorithms including the operation position, rotation direction, rotation angle, and tolerance of the stage rotation module corresponding to the model of the polarizing microscope device (110) and transmit them to the human robot (120) via the transmission unit (132). Additionally, if the additional observation condition is a polarizer rotation condition for verifying interference colors, the central hub system (130) can transmit additional polarizing microscope operation learning data and control algorithms including the reference position, rotation angle range, and shooting time of the polarizer rotation module to the human robot (120).

[0113] The human robot (120) can perform additional operations on the polarizing microscope device (110) based on additional polarizing microscope operation learning data and control algorithms received from the central hub system (130). For example, the human robot (120) can rotate the stage rotation module at predetermined angle intervals to check the extinction characteristics, and then cause the polarizing microscope device (110) to take additional mineral images. Additionally, the human robot (120) can rotate the polarizing plate rotation module by a predetermined angle to check the interference color, fine-tune the focus adjustment module to check the crystal boundary, and operate the magnification change module to check the enlargement of mineral particles, and then cause the polarizing microscope device (110) to acquire additional mineral images.

[0114] In one embodiment, the artificial intelligence analysis unit (140) may generate multiple additional observation conditions according to priority, rather than generating only a single additional observation condition, when the mineral identification reliability for the initial mineral image is below the reference reliability. For example, the artificial intelligence analysis unit (140) may first generate a stage rotation angle condition for verifying extinction characteristics, and if the analysis result of the additional mineral image according to the condition also does not reach the reference reliability, it may subsequently generate a polarizer rotation condition for verifying interference colors. In this way, the artificial intelligence analysis unit (140) may generate additional observation conditions step by step according to changes in mineral identification reliability, and the central hub system (130) may sequentially transmit additional polarizing microscope operation learning data and control algorithms corresponding to each step to the human robot (120).

[0115] In another embodiment, the artificial intelligence analysis unit (140) can determine the lack of observation information for each candidate mineral group and generate additional observation conditions differently. For example, if the candidate mineral group is a mineral group distinguished by a difference in extinction angle, the artificial intelligence analysis unit (140) can generate a stage rotation angle condition first, and if the candidate mineral group is a mineral group distinguished by a difference in interference color, it can generate a polarizer rotation condition first. In addition, if the candidate mineral group is a mineral group distinguished by crystal boundaries or particle shapes, it can generate a focus adjustment condition or a magnification change condition first. Accordingly, the mineral identification support system (100) can selectively supplement the observation conditions necessary for identification based on the analysis results of the artificial intelligence analysis unit (140), rather than performing uniform additional shooting.

[0116] In another embodiment, the central hub system (130) can combine additional observation conditions received from the artificial intelligence analysis unit (140) with model information of the polarizing microscope device (110) to convert them into specific operation commands or operation scenarios to be provided to the human robot (120). For example, even if the same stage rotation angle condition is generated, the polarizing microscope device (110) of the first model may have the stage rotation module positioned on the left, and the polarizing microscope device (110) of the second model may have the stage rotation module positioned on the right. In this case, the central hub system (130) can generate or select additional polarizing microscope operation learning data and control algorithms suitable for the model of the polarizing microscope device (110) using microscope model information stored in the model information storage unit (133) and operation mapping data generated by the operation mapping generation unit (134). Accordingly, the human robot (120) can perform additional operations while correcting the difference in the position of the operating part due to the difference in the model of the polarizing microscope device (110).

[0117] In another embodiment, the artificial intelligence analysis unit (140) can re-analyze the additional mineral image after it has been acquired to calculate an updated mineral identification reliability. If the updated mineral identification reliability is greater than or equal to the reference reliability, the artificial intelligence analysis unit (140) can output a final mineral identification result, and if the updated mineral identification reliability is still less than the reference reliability, the artificial intelligence analysis unit (140) can generate new additional observation conditions. Through such iterative analysis and additional observation structures, the mineral identification support system (100) can provide a dynamic closed-loop identification structure that automatically performs additional necessary microscope operations based on the analysis results, rather than a static analysis method that simply relies on the initial image.

[0118] In another embodiment, the data collection unit (135) can collect mineral identification reliability calculated by the artificial intelligence analysis unit (140), additional observation conditions, additional operation history performed by the human robot (120), and additional mineral images, and store them in the data storage unit (131). The algorithm update unit (136) can analyze which additional observation conditions contributed to improving the identification reliability of a specific mineral group based on the stored information. For example, if it is determined that the stage rotation angle condition has most significantly improved the mineral identification reliability for a specific mineral group, the algorithm update unit (136) can update the corresponding stage rotation angle condition as the priority additional observation condition for that mineral group. Accordingly, the central hub system (130) can continuously improve the method of generating additional observation conditions and providing additional polarizing microscope operation learning data based on actual operation data.

[0119] In another embodiment, the artificial intelligence analysis unit (140) may evaluate the quality of the mineral image itself. For example, the artificial intelligence analysis unit (140) may evaluate at least one of the brightness, contrast, focus sharpness, mineral particle occupancy within the field of view, noise level, or saturation area ratio of the mineral image, and if the image quality is below the reference quality, additional observation conditions may be generated separately from the mineral identification reliability. In this case, the additional observation conditions may include focus adjustment conditions for confirming crystal boundaries or magnification change conditions for confirming mineral particle magnification, and the central hub system (130) may provide additional polarizing microscope operation learning data and control algorithms corresponding to the conditions to the human robot (120). This configuration enables obtaining an image capable of analysis through additional shooting not only when the type of mineral is complex but also when the quality of the initial captured image is low.

[0120] In another embodiment, the artificial intelligence analysis unit (140) may be implemented as an internal component of the central hub system (130), or as a separate server, local analysis device, or cloud analysis device connected to the central hub system (130) via a communication network. When the artificial intelligence analysis unit (140) is implemented as an internal component of the central hub system (130), mineral image analysis, calculation of mineral identification reliability, generation of additional observation conditions, and selection of additional polarizing microscope operation learning data may be performed within a single integrated server. On the other hand, when the artificial intelligence analysis unit (140) is implemented as a separate analysis device, it may be implemented by receiving and analyzing mineral images from the polarizing microscope device (110) or the human robot (120), and then providing only additional observation conditions to the central hub system (130). These differences in implementation methods may be selected depending on the system installation environment, the level of data security requirements, the amount of analysis computation, and the network environment of the research institution.

[0121] And, FIG. 5 is a flowchart illustrating a mineral identification support method according to one embodiment of the present invention.

[0122] As illustrated in FIG. 5, a mineral identification support method according to one embodiment of the present invention may be a method in which a central hub system (130), a human robot (120), a polarizing microscope device (110), and an artificial intelligence analysis unit (140) interact with each other to automatically perform a polarizing microscope-based identification procedure for a mineral sample. The mineral identification support method according to this embodiment may be composed of a series of procedures in which a skilled researcher's mineral identification procedure is digitized, the operating characteristics of the polarizing microscope device (110) by model are reflected, the human robot (120) physically operates the polarizing microscope device (110) based on the data and algorithm, and the artificial intelligence analysis unit (140) analyzes the captured image to calculate the mineral identification result and reliability.

[0123] First, the central hub system (130) can learn the mineral identification process operation procedure of a skilled researcher and build corresponding data (S100). Here, the mineral identification process operation procedure of a skilled researcher may include a series of operation procedures such as rotating a polarizer, inserting or removing an analyzer, rotating a stage, adjusting focus, changing magnification, and setting the time of image capture, which are performed during the actual mineral identification process using a polarizing microscope device (110). For example, if a skilled researcher performs open Nicol observation and then performs orthogonal Nicol observation to identify a specific mineral sample, then checks the extinction characteristics while rotating the stage at regular angular intervals, and if necessary changes the focus position and magnification to capture additional images, the central hub system (130) can build these operation sequences, operation conditions, operation amounts, and capture times as learning data.

[0124] Next, the central hub system (130) can register model information of the polarizing microscope device (110) and construct control unit information and control mapping data (S110). Depending on the manufacturer or model, the polarizing microscope device (110) may differ in the position, shape, direction of operation, and driving method of the polarizer rotation module, analyzer insertion module, stage rotation module, focus adjustment module, and magnification change module. Accordingly, the central hub system (130) can register manufacturer information, model information, control unit placement information, control unit shape information, control unit driving method information, and allowable operation range information of the polarizing microscope device (110), and based on this, can construct control mapping data that can be used when a human robot (120) operates a specific model of the polarizing microscope device (110). This control mapping data may include reference position, approach direction, rotation direction, amount of operation, allowable error, and interference avoidance path for each control unit.

[0125] Subsequently, the central hub system (130) can generate and store mineral-specific operation scenarios, operation learning data, and control algorithms (S120). The mineral-specific operation scenarios may be data defining observation procedures required for mineral identification by mineral type or candidate mineral group. For example, for a specific mineral group, the stage rotation observation step may be emphasized because confirming extinction characteristics may be important, and for another mineral group, the polarizer rotation condition or orthogonal Nicol observation step may be emphasized because confirming interference colors may be important. The central hub system (130) can generate mineral-specific operation scenarios that include an open Nicol observation step, an orthogonal Nicol observation step, a stage rotation observation step, a extinction angle confirmation step, an interference color confirmation step, a focus adjustment step, a magnification change step, and an image capture step, depending on these mineral-specific characteristics.

[0126] The central hub system (130) can store generated operation learning data, control algorithms, model information, operation mapping data, and a scenario library in the data storage unit (131) (S130). The data storage unit (131) can store operation procedure data obtained during the mineral identification process of a skilled researcher and operation data by model of the polarizing microscope device (110) in correspondence with each other, and can be configured to search for data suitable for the task when an identification task for a specific mineral sample or candidate mineral group is requested. In addition, the data storage unit (131) can store a mineral-specific operation scenario library, and the mineral-specific operation scenario library may include the polarizing microscope operation sequence required for each type of mineral.

[0127] Subsequently, the central hub system (130) may receive an analysis request (S140). The analysis request may include at least one of the model information of the polarizing microscope device (110), mineral sample information, operation type information, candidate mineral group information, analysis purpose information, and required reliability information. For example, the analysis request may include the model name of the polarizing microscope device (110) installed at a specific research institution, the shape of the mineral sample to be analyzed, the operation type regarding whether it is thin section analysis or powder sample analysis, the expected candidate mineral group, and the standard reliability required for the final identification result.

[0128] The central hub system (130) can select appropriate operation learning data, control algorithms, operation mapping data, and operation scenarios based on the received analysis request (S150). At this stage, the central hub system (130) can select operation mapping data corresponding to the model information of the polarizing microscope device (110) and select mineral-specific operation scenarios corresponding to mineral sample information or candidate mineral group information. Additionally, the central hub system (130) can select a version of a control algorithm to be provided to the human robot (120) by considering together the identification information of the human robot (120), the model information of the polarizing microscope device (110), and the type of mineral identification task.

[0129] Next, the central hub system (130) can transmit selected data and algorithms to the human robot (120) (S160). The central hub system (130) can provide operation learning data, control algorithms, operation mapping data, and operation scenarios to the human robot (120) through the transmission unit (132). At this time, the transmission unit (132) may transmit data in batches through a communication network, or it may transmit only the necessary data sequentially according to the progress stage of the mineral identification process. For example, in the initial observation stage, the operation scenario required for open Nicol observation may be transmitted first, and then subsequent operation scenarios may be transmitted at the time when orthogonal Nicol observation or stage rotation observation is required.

[0130] The human robot (120) can receive operation learning data, control algorithms, operation mapping data, and operation scenarios from the central hub system (130) (S170). Based on the received data, the human robot (120) can identify the target module and operation sequence of the polarizing microscope device (110), and recognize the position of the operation part, the direction of operation, the amount of operation, and the tolerance corresponding to the model of the polarizing microscope device (110). Accordingly, the human robot (120) can perform operations corresponding to the actual equipment structure based on the model-specific operation mapping data provided by the central hub system (130), rather than using only fixed coordinates dependent on a specific polarizing microscope device (110).

[0131] The human robot (120) can perform recognition and calibration of the microscope control parts of the polarizing microscope device (110) (S210). At this stage, the human robot (120) can establish the position of the control parts of the polarizing microscope device (110), the device reference point, the robot reference coordinates, and the relative coordinate system between the polarizing microscope device (110) and the human robot (120). For example, the human robot (120) can use a robot vision module (121) to photograph the positions of the polarizer rotation module, analyzer insertion module, stage rotation module, focus adjustment module, and magnification change module of the polarizing microscope device (110), and use a control part recognition module (122) to recognize the position, shape, rotation axis direction, or movement direction of each control part. Subsequently, the human robot (120) can correct the actual control position by comparing the recognition results with the control mapping data provided from the central hub system (130).

[0132] The human robot (120) can perform automatic operation according to the operation scenario (S220). At this stage, the human robot (120) can perform at least one of rotating the polarizer, inserting or removing the analyzer, rotating the stage, adjusting the focus, and changing the magnification. The human robot (120) can operate the knob-type operating part, lever-type operating part, slide-type operating part, or button-type operating part of the polarizing microscope device (110) by gripping, rotating, pressing, pushing, or pulling using the end effector (123). Additionally, the human robot (120) can detect at least one of contact force, rotational resistance, movement resistance, and torque generated during the operation process through the force sensing module (124), and correct the gripping force, operation speed, or approach path based on the detection result.

[0133] The polarizing microscope device (110) can form various observation conditions according to the operation of the human robot (120). Specifically, the polarizing microscope device (110) can change polarization conditions by operating a polarizing plate rotation module (S221), insert an analyzer into or remove an analyzer from the optical path by operating an analyzer insertion module (S222), and change the observation orientation or rotation angle of the mineral sample by operating a stage rotation module (S223). In addition, the polarizing microscope device (110) can adjust the focus position for the surface, internal region, or crystal boundary of the mineral sample by operating a focus adjustment module (S224), and magnify and observe the entire structure or a specific particle region of the mineral sample by operating a magnification change module (S225). The operation of each of these modules can be achieved through physical operations performed directly by the human robot (120).

[0134] The human robot (120) can generate a conditional image capture request or trigger (S230). The conditional image capture request or trigger may be a signal to capture a mineral image at a specific point in time after a polarizer rotation condition, an analyzer insertion condition, a stage rotation condition, a focus adjustment condition, or a magnification change condition has been completed. For example, the human robot (120) can generate an image capture trigger at each rotation angle after sequentially operating the stage rotation module to positions of 0, 15, 30, and 45 degrees. Additionally, the human robot (120) can generate an image capture request when a condition is formed in which interference color verification is possible in an orthogonal Nicol state.

[0135] The polarizing microscope device (110) can capture a mineral image according to the conditional image capture request or trigger (S231). The mineral image captured by the polarizing microscope device (110) may be an image acquired under specific polarization conditions, a specific analyzer insertion state, a specific stage rotation angle, a specific focal position, or a specific magnification condition. Additionally, the mineral image may be a single image, or a set of multiple images acquired under different observation conditions for the same mineral sample.

[0136] A human robot (120) or a central hub system (130) can collect captured images and operation history (S240). The captured images may be mineral images acquired by a polarizing microscope device (110), and the operation history may include the history of polarizer rotation, the history of analyzer insertion or removal, the history of stage rotation, the history of focus adjustment, the history of magnification change, the time of image capture, and the time of execution of each operation step performed by the human robot (120). By collecting the captured images and operation history together in this way, it is possible to clearly identify under what observation conditions each image was acquired.

[0137] The artificial intelligence analysis unit (140) can perform mineral identification and reliability evaluation based on captured images (S250). The artificial intelligence analysis unit (140) can extract image features corresponding to color distribution, interference color patterns, crystal boundaries, grain shapes, texture patterns, extinction changes, birefringence characteristics, or optical anisotropy from the captured images, and can calculate a mineral candidate group or a final mineral identification result based on the extracted image features. Additionally, the artificial intelligence analysis unit (140) can calculate the mineral identification reliability of the calculated mineral identification result, and the mineral identification reliability can be expressed as a probability value, similarity value, reliability score, or degree of matching with reference data for each candidate mineral.

[0138] Subsequently, the artificial intelligence analysis unit (140) can determine whether the calculated mineral identification reliability is greater than or equal to the reference reliability (S260). The reference reliability may be a pre-set threshold value and may be set differently depending on the type of mineral, the purpose of analysis, the condition of the sample, the analysis policy of the research institution, or the type of work. For example, a relatively low reference reliability may be set for general educational analysis, and a higher reference reliability may be set for precision research analysis or industrial quality assessment.

[0139] If the mineral identification reliability is greater than or equal to the standard reliability, the human robot (120), the central hub system (130), or the artificial intelligence analysis unit (140) may transmit the final result and operation history (S270). The final result may include the identified mineral name, candidate mineral group, mineral identification reliability, the image used for analysis, shooting conditions, and operation history. The operation history is data indicating which polarizing microscope operation procedure was used to obtain the result, and can be used for subsequent verification, researcher review, algorithm updates, or the generation of analysis reports.

[0140] On the other hand, if the reliability of mineral identification is below the standard reliability, the artificial intelligence analysis unit (140) can generate additional observation conditions, and the human robot (120) can perform automatic operations again according to the additional observation conditions. For example, if the artificial intelligence analysis unit (140) determines that confirmation of extinction characteristics is insufficient, it can generate a stage rotation angle condition as an additional observation condition, and if it determines that confirmation of interference colors is insufficient, it can generate a polarizer rotation condition as an additional observation condition. In addition, if it determines that crystal boundaries are unclear, it can generate a focus adjustment condition as an additional observation condition, and if confirmation of the detailed shape of mineral particles is required, it can generate a magnification change condition as an additional observation condition. In this case, the central hub system (130) can provide additional polarizing microscope operation learning data and control algorithms corresponding to the additional observation conditions to the human robot (120), and the human robot (120) can perform additional operations on the polarizing microscope device (110) based on this and then acquire additional mineral images.

[0141] The central hub system (130) can receive operation history, images, and analysis results (S180). The central hub system (130) can receive operation history, captured images, mineral identification reliability, additional observation conditions, and final identification results for each stage, not only when a final result is obtained, but also when additional observation is performed due to a result below the reference reliability. This data is not merely result storage data, but can be used as training data for improving mineral-specific operation scenarios and updating control algorithms in the future.

[0142] The central hub system (130) can update algorithms and improve performance based on collected data (S190). For example, the algorithm update unit (136) can analyze which operation sequence, operation angle, operation amount, or shooting time improved the reliability of mineral identification for a specific mineral group. If the stage rotation angle condition repeatedly contributed to the improvement of reliability in a specific mineral group, the algorithm update unit (136) can extract the corresponding stage rotation angle condition as a valid operation pattern and reflect it in the mineral-specific operation scenario library. Additionally, if operation failures or errors repeatedly occur in the model of a specific polarizing microscope device (110), the algorithm update unit (136) can modify the operation mapping data or tolerance range for the corresponding model.

[0143] The central hub system (130) can redistribute updated algorithms and data (S200). The updated algorithms and data can be stored in the data storage unit (131) and provided to human robots (120) of the same research institution or other research institutions through the delivery unit (132). If the central hub system (130) is implemented as a cloud server, it can redistribute improved algorithms to multiple human robots (120) based on operation history, captured images, and analysis results collected from multiple research institutions or multiple analysis equipment. Accordingly, even if the human robots (120) of each research institution use different polarizing microscope devices (110), they can perform operations related to the mineral identification process by applying the latest operation learning data and control algorithms updated by the central hub system (130).

[0144] In one embodiment, the mineral identification support method of FIG. 5 can be performed for a single analysis task. In this case, the central hub system (130) provides operation learning data, control algorithms, operation mapping data, and operation scenarios to one polarizing microscope device (110) and one human robot (120), and the human robot (120) can automatically perform polarizing microscope operations on a mineral sample using the data. Subsequently, the polarizing microscope device (110) captures mineral images according to conditions, and the artificial intelligence analysis unit (140) analyzes the captured images to calculate mineral identification results and reliability.

[0145] In another embodiment, the mineral identification support method of FIG. 5 may be performed in parallel for a plurality of research institutions or a plurality of polarizing microscope devices (110). In this case, the central hub system (130) may select and provide different operation mapping data and control algorithms based on the model information of the polarizing microscope device (110) of each research institution and the identification information of the human robot (120). For example, the human robot (120) of the first research institution may receive operation data corresponding to the first model of the polarizing microscope device (110), and the human robot (120) of the second research institution may receive operation data corresponding to the second model of the polarizing microscope device (110). Accordingly, the mineral identification support method of FIG. 5 is not limited to a specific model of the polarizing microscope device (110) but can be applied to polarizing microscope devices (110) of various manufacturers and models.

[0146] In another embodiment, the mineral identification support method of FIG. 5 can be implemented in a closed-loop manner that repeatedly performs additional observations based on the reliability evaluation results of the artificial intelligence analysis unit (140). That is, if the artificial intelligence analysis unit (140) produces a result below the reference reliability for the initial captured image, the system automatically generates additional observation conditions, the central hub system (130) provides additional operation data corresponding to the conditions to the human robot (120), and the human robot (120) can perform additional operations on the polarizing microscope device (110). Subsequently, the additional mineral image is analyzed again by the artificial intelligence analysis unit (140), and the reliability judgment can be repeated. Due to this closed-loop structure, the mineral identification support method of FIG. 5 can be implemented not as a static analysis method that simply relies on the initial image, but as a dynamic mineral identification method that actively supplements insufficient observation conditions based on the analysis results.

[0147] In another embodiment, the mineral identification support method of FIG. 5 can be applied to educational mineral analysis, research mineral analysis, resource exploration sample analysis, industrial raw mineral analysis, rock thin section analysis, or automated laboratory analysis. For example, in an educational environment, a human robot (120) can reproduce a standardized mineral identification procedure without the need for a student or novice to directly perform the operation procedure of a skilled researcher, thereby reducing variations in observation conditions. In a research environment, repetitive polarizing microscope operations for multiple mineral samples can be automated, thereby reducing analysis time and the workload of the researcher. In an industrial environment, consistent operation conditions can be maintained in repeated analysis of the same mineral sample group, thereby improving the reproducibility of the analysis results.

[0148] Although various preferred embodiments of the present invention have been described above with some examples, the descriptions of various embodiments described in the "Specific details for carrying out the invention" section are merely illustrative, and those skilled in the art to which the present invention pertains will understand that the present invention can be modified in various ways or equivalent embodiments can be carried out based on the above description.

[0149] In addition, since the present invention can be implemented in various other forms, the present invention is not limited by the description above. The above description is provided merely to make the disclosure of the present invention complete and to fully inform those skilled in the art of the scope of the present invention, and it should be understood that the present invention is defined only by each claim of the claims. Explanation of the symbols

[0150] 100: Mineral Identification Support System 110: Polarizing microscope device 120: Humanoid Robot 130: Central Hub System 140 : Artificial Intelligence Analysis Department

Claims

Claim 1 A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system for identifying the type of mineral contained in a mineral sample, comprising: a polarizing microscope device for observing the mineral sample; a human robot configured to physically operate at least one of a polarizing plate rotation module, an analyzer insertion module, a stage rotation module, a focus adjustment module, and a magnification change module of the polarizing microscope device; and a central hub system that integrally manages polarizing microscope operation learning data and control algorithms corresponding to a plurality of polarizing microscope models and provides polarizing microscope operation learning data and control algorithms corresponding to the model of the polarizing microscope device to the human robot, wherein the central hub system comprises: a data storage unit that stores polarizing microscope operation learning data and control algorithms for polarizing microscope operation procedures performed during the mineral identification process of a skilled researcher; A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system, comprising a transmission unit that transmits polarizing microscope operation learning data and control algorithms corresponding to a model of the polarizing microscope device to the human robot, wherein the human robot is configured to recognize the position or shape of the operation unit of the polarizing microscope device based on the polarizing microscope operation learning data and control algorithms received from the central hub system, and to automatically perform operations related to the mineral identification process for the polarizing microscope device. Claim 2 A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to claim 1, wherein the human robot comprises: a robot vision module for photographing the operating part of the polarizing microscope device; and an operating part recognition module that recognizes the position of at least one of the polarizer rotation module, analyzer insertion module, stage rotation module, focus adjustment module, and magnification change module based on the operating part image obtained by the robot vision module. Claim 3 A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to claim 1, wherein the human robot includes an end effector for gripping or rotating the operating part of the polarizing microscope device, and the end effector is configured to be replaceable or have a variable gripping width corresponding to at least one of a knob-type operating part, a lever-type operating part, a slide-type operating part, and a button-type operating part. Claim 4 A physical AI-based human-robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to claim 1, wherein the polarizing microscope operation learning data includes at least one of the polarizing plate rotation angle, whether an analyzer is inserted, stage rotation angle, focus adjustment amount, magnification change conditions, and image capture time performed by a skilled researcher during the mineral identification process. Claim 5 A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to claim 1, wherein the central hub system further comprises a model information storage unit that stores microscope model information including manufacturer information, model information, control unit placement information, and control unit driving method information of the polarizing microscope device. Claim 6 In claim 5, the central hub system further includes a control mapping generation unit that generates control mapping data including a reference position, control direction, control amount, and tolerance for each control part of the polarizing microscope device based on the microscope model information, and the transmission unit transmits the control mapping data, the polarizing microscope control learning data, and a control algorithm together to the human robot, and the human robot performs a control related to the mineral identification process after correcting the difference in control part positions according to the model difference of the polarizing microscope device based on the control mapping data, characterized by a physical AI-based human robot polarizing microscope automatic control and central hub-linked mineral identification support system. Claim 7 A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to claim 1, wherein the human robot includes a force sensing module that detects at least one of contact force, rotational resistance, movement resistance, and torque generated during the process of operating the operating part of the polarizing microscope device, and adjusts the gripping force or operating speed for the operating part based on the detection result of the force sensing module. Claim 8 A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to claim 1, wherein the system further includes an artificial intelligence analysis unit that analyzes a mineral image captured by the polarizing microscope device to calculate a mineral identification reliability, wherein the artificial intelligence analysis unit generates additional observation conditions when the mineral identification reliability is less than a reference reliability, wherein the central hub system transmits additional polarizing microscope operation learning data and a control algorithm corresponding to the additional observation conditions to the human robot, and wherein the human robot performs additional operations on the polarizing microscope device based on the additional polarizing microscope operation learning data and the control algorithm, after which an additional mineral image is acquired. Claim 9 A physical AI-based human-robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to claim 8, wherein the additional observation conditions include at least one of a stage rotation angle condition for confirming extinction characteristics, a polarizer rotation condition for confirming interference colors, a focus adjustment condition for confirming crystal boundaries, and a magnification change condition for confirming mineral grain magnification. Claim 10 A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to claim 1, wherein the central hub system further comprises a data collection unit that collects polarizing microscope operation history and captured images performed during the mineral identification process from multiple research institutions or analysis equipment. Claim 11 In claim 10, the above central hub system further comprises an algorithm update unit that updates the polarizing microscope operation learning data and control algorithm using the polarizing microscope operation history and captured images collected by the data collection unit, thereby providing a physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system. Claim 12 In claim 11, the algorithm update unit extracts the operation sequence, operation angle, operation amount, and shooting time that improve mineral identification accuracy from the polarizing microscope operation history as a valid operation pattern based on whether the mineral identification result for the captured image is correct or reliable, and updates the valid operation pattern as a polarizing microscope operation scenario for each mineral type, thereby providing a physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system. Claim 13 A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to claim 1, wherein the central hub system stores a mineral-specific operation scenario library including a polarizing microscope operation sequence required for each mineral type, and the mineral-specific operation scenario library includes at least two steps among an open Nicol observation step, an orthogonal Nicol observation step, a stage rotation observation step, a light extinction angle verification step, and an interference color verification step, and the transmission unit transmits a mineral-specific operation scenario corresponding to a candidate mineral group of a mineral sample to be analyzed to the human robot. Claim 14 A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to claim 1, wherein the human robot performs a calibration operation to identify a reference position of the polarizing microscope device before performing an operation on the polarizing microscope device, and establishes a relative coordinate system between the polarizing microscope device and the human robot based on the result of the calibration operation. Claim 15 A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to claim 14, wherein the human robot calculates an approach path for each operating part of the polarizing microscope device based on the relative coordinate system, and the approach path is set to avoid interference with at least one of the eyepiece, objective lens, sample stage, and imaging module of the polarizing microscope device. Claim 16 A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system according to claim 1, wherein the central hub system further includes an algorithm selection unit for selecting a version of polarizing microscope operation learning data and a control algorithm to be provided to the human robot based on the identification information of the human robot, the model information of the polarizing microscope device, and the type of mineral identification task. Claim 17 A physical AI-based human robot polarizing microscope automatic operation and central hub-linked mineral identification support system, characterized in that, in claim 1, the central hub system is implemented as a cloud server, the transmission unit provides polarizing microscope operation learning data and control algorithms to each of a plurality of human robots via a communication network, and each of the plurality of human robots performs operations related to the mineral identification process based on polarizing microscope operation learning data and control algorithms corresponding to models of different polarizing microscope devices.

Images