Dissolution determination apparatus and its operating method

Abstract

A dissolution determination apparatus and its operating method are provided. The dissolution determination apparatus may include: a reader configured to read identification information of a container holding a target sample; a drive module including at least one body configured to open the container and move the container to a capture position; an image acquisition module configured to acquire at least one image of the target sample at the capture position when the container is opened; and at least one processor configured to automatically determine whether the target sample has dissolved by analyzing at least one image of the target sample based on the identification information of the container.

Description

[0001] Cross-references to related applications

[0002] This application claims priority to Korean Patent Application No. 10-2024-0201242, filed on December 30, 2024, with the Korean Intellectual Property Office, the disclosure of which is incorporated herein by reference in its entirety. Technical Field

[0003] The methods and apparatus consistent with the embodiments relate to a dissolution determination apparatus and its operating method. Background Technology

[0004] To measure dissolution, methods such as measuring optical density using a spectrophotometer, or measuring turbidity using a nephelometer or turbidimeter, can be used. All of the above methods utilize light scattering, and therefore, real-time measurement of dissolution is not possible. These methods can measure the degree to which a solute dissolves in a solution rather than directly sensing dissolution. These methods directly measure the solvent or solute, or use absorbance density and spectroscopy to measure the solvent or solute to monitor the degree of dissolution, and are therefore limited in monitoring the entire region of interest (ROI). Furthermore, methods using computer vision to sense dissolution may have limitations in scalability because these methods measure images in a fixed direction, and the purpose of the method is to measure dissolution. Summary of the Invention

[0005] One or more embodiments of this disclosure can solve at least the problems and / or disadvantages described above, as well as other disadvantages not described above. Furthermore, embodiments of this disclosure do not need to overcome the aforementioned disadvantages, and embodiments of this disclosure may not overcome any of the aforementioned problems.

[0006] According to an aspect of this disclosure, a dissolution determination apparatus may include: a reader configured to read identification information of a container holding a target sample; a drive module including at least one body configured to open the container and move the container to a capture position; an image acquisition module configured to acquire at least one image of the target sample at the capture position when the container is opened; and at least one processor configured to automatically determine whether the target sample has been dissolved by analyzing at least one image of the target sample based on the identification information of the container.

[0007] The reader can be further configured to sense the location of the container.

[0008] At least one body of the drive module may include at least one of the following: a gripper configured to grip a container; a robot configured to transfer the container to a capture position along a transfer path; an actuator configured to adjust at least one of the container's height and horizontal position; and a capper configured to open the container.

[0009] The image acquisition module includes at least one of the following: a first camera configured to capture a first image of the upper part of the target sample when the target sample is in the container and the container is open; a second camera configured to capture a second image of the side of the target sample when the target sample is in the container and the container is open; and a third camera configured to capture a third image of the lower part of the target sample when the target sample is in the container, wherein at least one processor may be configured to automatically determine whether the target sample has been dissolved by analyzing at least one of the first image, the second image and the third image of the target sample based on container identification information.

[0010] At least one processor may be further configured to: extract a region of interest (ROI) from at least one image of the target sample; calculate a noise level corresponding to the ROI by frequency-based filtering of the ROI; detect particles of undissolved solute included in the ROI; and determine whether the target sample has been dissolved based on the noise level and the particles of undissolved solute.

[0011] At least one processor may be further configured to detect particles of undissolved solute using at least one of a bandpass filter, an adaptive threshold, and a nonlinear filter.

[0012] At least one processor may be further configured to: extract features from the ROI; obtain analytical results by analyzing the degree of solubility of the target sample via applying the features to a neural network trained based on an analytical algorithm; and generate control signals based on the analytical results.

[0013] The analysis algorithm can be configured to analyze at least one of the following: whether the target sample is completely dissolved, the opacity of the target sample, undissolved solute particles in the target sample, and residues around the container.

[0014] At least one processor may be further configured to: determine dissolution conditions corresponding to the target sample based on the container's identification information; and, based on the determination that the target sample has not been completely dissolved, generate a control signal for accelerating the dissolution of the target sample according to the dissolution conditions.

[0015] Dissolution conditions may include at least one of the following: the type of solvent used to dissolve the target sample; the amount of solvent; the type of catalyst used to dissolve the target sample; the amount of catalyst, temperature, humidity, or pressure; and the number of times the container is agitated.

[0016] According to an aspect of this disclosure, a method performed by a dissolution determination device may include: reading identification information of a container containing a target sample; opening the container; moving the container to a capture position; acquiring at least one image of the target sample at the capture position while the container contains the target sample and the container is open; and automatically determining whether the target sample has been dissolved by analyzing at least one image of the target sample based on the identification information of the container.

[0017] Opening a container may include: adjusting at least one of the container's height and position; and opening the container after adjusting at least one of the container's height and position.

[0018] Obtaining at least one image of the target sample may include: capturing a first image of the upper part of the target sample when the container contains the target sample and is open; capturing a second image of the side of the target sample when the container contains the target sample and is open; and capturing a third image of the lower part of the target sample when the container contains the target sample.

[0019] Automatically determining whether a target sample has been dissolved may include: extracting a region of interest (ROI) from at least one image of the target sample; and determining whether the target sample has been dissolved based on frequency-based filtering of the ROI.

[0020] Determining whether a target sample has been dissolved based on frequency-based filtering may include: using image entropy to calculate the noise level corresponding to the ROI; detecting undissolved solute particles included in the ROI by frequency-based filtering with respect to the ROI; and determining whether the target sample has been dissolved based on the noise level and the undissolved solute particles.

[0021] Detection of undissolved solute particles may include using at least one of a bandpass filter, an adaptive threshold, and a nonlinear filter.

[0022] Automatically determining whether a target sample has dissolved may include: extracting features from the ROI; obtaining analytical results by analyzing the degree of dissolution of the target sample through applying the features to a neural network trained based on an analytical algorithm; and generating control signals based on the analytical results.

[0023] The analysis algorithm can analyze at least one of the following: whether the target sample is completely dissolved, the opacity of the target sample, undissolved solute particles in the target sample, and residues around the container.

[0024] Automatically determining whether a target sample has been dissolved may include: determining the dissolution conditions corresponding to the target sample based on the container's identification information; and generating a control signal to accelerate the dissolution of the target sample based on the dissolution conditions, based on the determination that the target sample has not been completely dissolved.

[0025] According to aspects of this disclosure, dissolution conditions may include at least one of the following: the type of solvent used for dissolving the target sample; the amount of solvent; the type of catalyst used for dissolving the target sample; the amount of catalyst, temperature, humidity, or pressure; and the number of times the container is agitated.

[0026] Additional aspects of embodiments of this disclosure will be set forth in part in the description which follows, and will be apparent in part from the description, or may be learned by practice of this disclosure. Attached Figure Description

[0027] The above and / or other aspects will become more apparent from the description of certain embodiments with reference to the accompanying drawings, in which:

[0028] Figure 1 This is a block diagram of a dissolution determination apparatus according to an embodiment;

[0029] Figure 2A This is a front view of the dissolution determining apparatus according to an embodiment;

[0030] Figure 2B This is a bird's-eye view of the dissolution determination apparatus according to an embodiment;

[0031] Figure 2C This is a side view of the dissolution determining apparatus according to an embodiment;

[0032] Figure 3 This is a flowchart illustrating a method for operating the dissolution determination apparatus according to an embodiment;

[0033] Figure 4 This is a flowchart illustrating a method for operating the dissolution determination apparatus according to an embodiment;

[0034] Figure 5 This is a schematic diagram illustrating a method for extracting a region of interest (ROI) from an image of a target sample according to an embodiment;

[0035] Figure 6 This is a schematic diagram illustrating a method for calculating the noise level corresponding to the ROI according to an embodiment;

[0036] Figure 7 This is a schematic diagram illustrating a method for determining whether a target sample has dissolved according to an embodiment; and

[0037] Figure 8 This is a flowchart illustrating a method for operating the dissolution determination apparatus according to an embodiment. Detailed Implementation

[0038] The following detailed structural or functional description is provided to explain non-limiting exemplary embodiments of this disclosure, and embodiments of this disclosure may include various changes and modifications. Therefore, embodiments of this disclosure are not limited to the exemplary embodiments and should be understood to include all changes, equivalents, and substitutions within the spirit and scope of this disclosure.

[0039] Although terms such as "first" and "second" are used to describe various components, components are not limited to these terms. These terms may only be used to distinguish one component from another. For example, the first component may be referred to as the second component, or similarly, the second component may be referred to as the first component.

[0040] It should be noted that if a component is described as “connecting,” “coupled,” or “joining” to another component, a third component may be “connected,” “coupled,” or “joined” between the first and second components, although the first component may be directly connected, coupled, or joined to the second component.

[0041] Unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well. It will be further understood that, when used herein, the terms “comprises / comprising” and / or “includes / including” specify the presence of the declared features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0042] Unless otherwise defined, all terms used herein, including technical or scientific terms, shall have the same meaning as commonly understood by one of ordinary skill in the art to which the embodiments pertain. Terms, such as those defined in common dictionaries, shall be interpreted as having a meaning matching the contextual meaning in the relevant field and shall not be interpreted as having an ideal or overly formal meaning, unless otherwise defined herein.

[0043] In the following, non-limiting exemplary embodiments of this disclosure will be described in detail with reference to the accompanying drawings. When describing exemplary embodiments with reference to the accompanying drawings, the same reference numerals denote the same elements, and repeated descriptions associated with those elements may be omitted.

[0044] Figure 1 This is a block diagram of a dissolution determination apparatus according to an embodiment. (See reference) Figure 1The dissolution determination apparatus (hereinafter referred to as "determination apparatus 100") according to the embodiment may include a reader 110, a driving module 130, an image acquisition module 150, and a processor 170. The determination apparatus 100 may further include a memory 190.

[0045] The determining device 100 can be a measuring device that automatically determines whether a compound has dissolved based on optics and computer vision. The determining device 100 can automate the entire process of measuring dissolution and determining whether a compound has dissolved, allowing the entire process to be performed without human intervention. The determining device 100 can operate in a closed-loop manner through feedback from other systems before and after the operation.

[0046] Reader 110 can read the container holding the target sample (e.g., Figure 2A The identification information of the container 205. The “target sample” can be a sample used as a measurement target to determine whether a sample has dissolved or not. The target sample can include at least one of a solid sample, a liquid sample, and a mixture of solid and liquid samples. Liquid samples can be in various forms such as liquids, pastes, sludge, and viscous oil. Solid samples can be in various forms such as powders, granules, pellets, films, and / or fibers. When the target sample is a solid sample such as a film, the determining device 100 can, for example, analyze not only the dissolution of the target sample but also the homogeneity of the target sample, or whether the turbidity of the target sample increases due to the reaction.

[0047] The container can be a vial for storing liquid drugs, reagents, powders, and / or pills, or a flask for containing samples, but is not limited to these. When the target sample is a liquid sample, the container can have various shapes that can hold the liquid sample without spilling. When the target sample is a liquid sample, the container can include, for example, a transparent flask or a cuvette. When the target sample is a solid sample, the container can be a flat container in which a solid sample can be placed. When the target sample is a solid sample, the container can include, for example, a plate or a slide. The following description is based on the case where the target sample is a liquid sample for ease of description. However, the description does not exclude the case where the target sample is a solid sample.

[0048] In the following text, even if not described separately, "container of the target sample" can be understood as a container holding a solution in which the target sample (solute) is dissolved in a solvent. Hereinafter, target sample and solute are used interchangeably.

[0049] Reader 110 can read at least one of the container's location and the container's identification information. Reader 110 may be, for example, a Quick Response (QR) reader that reads a QR code attached to the container, or a tag reader that reads a tag attached to the container, but is not limited thereto.

[0050] The drive module 130 can open the container by moving the container holding the target sample to or toward the capture position. For example, the drive module 130 can remove the lid from the container. According to some embodiments, the lid can be configured to attach to the upper end of the container to close it, and to be removed from the upper end of the container to open it. The drive module 130 can fix and / or move the container holding the target sample according to control signals from the processor 170. Here, "movement" can be understood as, for example, rotation, translation, and / or displacement. According to some embodiments, the drive module 130 can open the container at the capture position or at a position before the container reaches the capture position. According to some embodiments, the capture position can be the position where at least one image of the container is captured by the image acquisition module 150.

[0051] The drive module 130 can perform rotational drive on the container and / or the equipment used to fix the container according to the control signals of the processor 170. The drive module 130 can rotate the container clockwise or counterclockwise according to the control signals of the processor 170. The drive module 130 can change the posture of the container and / or the equipment used to fix the container according to the control signals of the processor 170.

[0052] The drive module 130 may include a gripper (e.g., Figure 2A The gripper 220), the robot (e.g., Figure 2A The gantry robot 230), actuators (e.g., Figure 2C The actuator 250) and the capper (e.g., Figure 2A At least one of the capping devices 240.

[0053] The clamp can hold the container so that it does not shake.

[0054] A robot can transfer a container to a capture position along a transfer path. The robot can be, for example, a gantry robot that moves along the X and Y axes. The gantry robot can be a Cartesian coordinate robot that performs tasks by moving from a fixed position along orthogonal axes represented as the X, Y, and / or Z axes. Here, the X-axis can be a first horizontal axis, the Y-axis can be a second horizontal axis perpendicular to the first horizontal axis, and the Z-axis can be a vertical axis perpendicular to both the first and second horizontal axes. The length of each axis can correspond to the operating range of the gantry robot. The gantry robot can use a device capable of gripping parts (e.g., a gripper) to transfer the container to the capture position along the transfer path.

[0055] The actuator can adjust at least one of the height and horizontal position of the container holding the target sample. The actuator can adjust the height and / or horizontal position of the container for use with a capping device.

[0056] The capper can close or open a container by attaching or removing the cap, wherein at least one of the container's height and horizontal position is adjusted.

[0057] The image acquisition module 150 can acquire an image of the target sample when the container holding the target sample is opened (e.g., the lid is removed from the container). The image acquisition module 150 can acquire an image of the target sample by changing the acquisition angle according to the control signal of the processor 170.

[0058] Image acquisition module 150 may include: a camera (e.g., a first camera) that captures images of the target sample in various directions. Figure 2C The first camera 251), the second camera (e.g., Figure 2C The second camera 253), and / or the third camera (e.g., Figure 2C At least one of the third camera 255), used as a light source for camera capture (e.g., Figure 2C (bottom light source 257 and / or side light source 259), and reflector.

[0059] The processor 170 can acquire at least two images obtained by capturing the target sample by changing the capture position or by rotating the camera of the image acquisition module 150 to change the displayed image. The at least two images are then compared to determine whether the target sample has dissolved. In this embodiment, by capturing the target sample contained in the container in multiple directions, errors that may occur when measuring images in one direction can be avoided.

[0060] Alternatively, the processor 170 may acquire an image of the interior of the container, i.e., an image of the target sample, via the image acquisition module 150 while or after the drive module 130 rotates the container up, down, left, and / or right.

[0061] The processor 170 can determine with greater accuracy whether a target sample has dissolved by performing image processing on the image acquired by the image acquisition module 150 using an image processing algorithm. The image processing algorithm can detect the size and number of particles in the undissolved solute from the captured image of the target sample. Moreover, the image processing algorithm can measure the solubility, turbidity, and amount of remaining undissolved solute in the solution in real time.

[0062] For example, when the processor 170 determines via an image processing algorithm that the solute is completely dissolved in the solution contained in the container, i.e., completely dissolved, the processor 170 can output "Pass". When the processor 170 determines via an image processing algorithm that undissolved solute exists in the solution contained in the container, or that the target sample is in a supersaturated state (crystalline state) or a turbid state, in other words, when the solution contained in the container is in an undissolved state, the processor 170 can output "Fail".

[0063] Furthermore, the processor 170 can classify the solubility of the target sample into several states and output the solubility state as the analysis result. For example, when the target sample is in a turbid state, the processor 170 can output "Failure #1" as the analysis result. Conversely, when the target sample is in a supersaturated state (crystalline state), the processor 170 can output "Failure #2" as the analysis result. When the processor 170 outputs analysis results by classifying them into pass, failure #1, and failure #2 based on the solubility of the target sample, the user can intuitively determine the solubility of the target sample.

[0064] Processor 170 can automatically determine whether a target sample has been dissolved by analyzing an image of the target sample based on the identification information of the container holding the target sample. In addition to information that can identify the target sample (such as an ID or QR code), the identification information of the container holding the target sample may include information about the name of the solvent contained in the target sample, the name of the solute, the name of the catalyst, and their volume. Processor 170 can automatically determine whether the target sample has been dissolved by identifying the state of the solution in the image of the target sample, whether it contains undissolved solute or a solution containing completely dissolved solute, based on the identification information.

[0065] More specifically, processor 170 can extract regions of interest (ROIs) from images of the target sample. See below for reference. Figure 5 A more detailed description of the method for detecting ROI using a deterministic device.

[0066] Processor 170 can calculate the noise level corresponding to the ROI through frequency-based filtering. See below for reference. Figure 6 The method for calculating the noise level corresponding to the ROI by a determining device is described in more detail.

[0067] Processor 170 can detect particles of undissolved solute contained in the ROI. Processor 170 can use at least one of a bandpass filter, an adaptive threshold, and a nonlinear filter to detect particles of undissolved solute. Processor 170 can determine whether the target sample has been dissolved based on the noise level and the presence of particles of undissolved solute.

[0068] Furthermore, the processor 170 can extract features from the ROI. Based on the extracted features, the processor 170 can analyze the degree of solubility of the target sample as, for example, a completely dissolved state, a supersaturated state, and a turbid state. The processor 170 can analyze the degree of solubility of the target sample, for example, by inputting the features extracted from the ROI into an analysis algorithm, or by applying the extracted features to a neural network trained based on the analysis algorithm. The analysis algorithm can analyze at least one of the following: whether the target sample is completely dissolved, the opacity of the target sample, undissolved solute particles in the target sample, and residues around the container (e.g., bubbles, water droplets, dust, etc.).

[0069] Processor 170 can generate control signals based on the analysis results. These control signals may include, for example, control signals for fixing or moving the container using drive module 130 based on the analysis results. Alternatively, processor 170 may generate different control signals depending on, for example, whether the analysis result is successful or unsuccessful. For instance, when the analysis result is successful (pass), processor 170 may generate control signals to continue the process (e.g., filtration) after determining whether the target sample has dissolved. Alternatively, when the analysis result is unsuccessful (failure), processor 170 may generate control signals for adding solvent to the container, adjusting the container temperature, adjusting the agitation speed of drive module 130, or increasing the reaction time of the target sample.

[0070] Processor 170 can determine the dissolution conditions corresponding to the target sample based on the container's identification information. Processor 170 can generate a control signal to accelerate the dissolution of the target sample based on whether the target sample has been completely dissolved. The dissolution conditions may include at least one of the following: the type of solvent used to dissolve the target sample; the amount of solvent; the type of catalyst used to dissolve the target sample; the amount of catalyst, temperature, humidity, and / or pressure; and the number of agitations in the container holding the target sample, but are not limited thereto.

[0071] Processor 170 may include a processor module of a user terminal such as a personal computer (PC), laptop, or tablet computer. Alternatively, processor 170 may drive a neural network-based analysis model by executing at least one program stored in memory 190. According to some embodiments, processor 170 may include one or more processors. According to some embodiments, memory 190 may store at least one program, and the program, when executed by processor 170, may be configured to cause processor 170 to perform the functions of the program. For example, the program may be configured to cause processor 170 to perform the following references. Figures 3 to 8 The operation of the described method.

[0072] The processor 170 can use a pre-trained neural network-based analysis model stored in the memory 190 to analyze the solubility of the target sample, the presence of undissolved solute particles in the target sample, and / or the presence of residues around the container from images acquired by the image acquisition module 150, and output the analysis results. The analysis model can be trained based on various analysis algorithms. The analysis model according to the embodiment can be implemented by various types of devices, such as, for example, PCs, server devices, mobile devices, and embedded devices. The analysis model can be implemented by an automatic material retrieval device that performs image recognition, image classification, etc., using a neural network, but is not limited thereto. Furthermore, the analysis device can be a dedicated hardware (HW) accelerator installed in the aforementioned devices, or it can be an HW accelerator such as a neural processing unit (NPU), tensor processing unit (TPU), or neural engine as a dedicated module for operating neural networks, but is not limited thereto.

[0073] Memory 190 can store at least one program. Additionally, memory 190 can store various information generated during processing by processor 170. Memory 190 can store a neural network trained based on an analysis algorithm. Furthermore, memory 190 can store various data and programs. Memory 190 may include, for example, volatile or non-volatile memory. Memory 190 may include a high-capacity storage medium such as a hard disk to store various data.

[0074] In addition, Figure 1 In addition, processor 170 can perform the following operations, as will be shown in Figure 2. Figure 8At least one method or a corresponding scheme described herein. Processor 170 may be a hardware-implemented dissolution determination device, dissolution measurement device, or analysis device having physically structured circuitry for performing the desired operation. The desired operation may be implemented, for example, by code or instructions included in a program. The determination device 100, which may be implemented in hardware, may include, for example, a microprocessor, a central processing unit (CPU), a graphics processing unit (GPU), a processor core, a multi-core processor, a multiprocessor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), and a neural processing unit (NPU).

[0075] The determining device 100 can be used to monitor the dissolution state of compounds, determine the degree of dissolution of biological materials, and / or retrieve conditions for the separation and analysis of materials, and / or be used as a water level sensor. Additionally, the determining device 100 can be used in the development of catalysts for the synthesis of cosmetics and fuel cells, which may require high capacity and high efficiency.

[0076] Figure 2A This is a front view of the dissolution determination apparatus according to an embodiment. Figure 2B This is a bird's-eye view of the dissolution determination apparatus according to an embodiment. Figure 2C This is a side view of the dissolution determination apparatus according to an embodiment.

[0077] refer to Figure 2A , 2B According to the embodiments, the operation of the determining device 100 can mainly include four stages: an input / output (I / O) stage, a capping stage, an image capture stage, and an image processing stage.

[0078] The I / O phase (“first phase”) can be a process of identifying the container to determine whether the target sample has been dissolved. In the I / O phase, the determining device 100 can verify the location of the container 205 (e.g., a formulation vial) containing the dissolved sample to be analyzed and the container’s identification information (ID) via a reader (e.g., a QR reader 210) before and after measuring whether the target sample has been dissolved.

[0079] The capping stage (“second stage”) can be a process of moving the container containing the target sample to the capture position and opening the container (e.g., removing the cap from the container). During the capping stage, the determining device 100 can hold the container 205 via a gripper 220 and can open / close (uncap / cap) the container via a capper 240, which can be a device for opening and closing the container 205 by removing the cap from the container 205 and attaching the cap to the container 205, respectively. The gripper 220 can include a polytetrafluoroethylene (PTFE) material or a PTFE-coated tip such that the position of the container 205 is not changed by friction or adhesive forces when the container 205 is placed.

[0080] Additionally, the determining device 100 can adjust the height of the container 205 via actuator 250 during the capping stage, and then transfer the container 205 along transfer path 215 via gantry robot 230. Actuator 250 may include two actuators configured to move the container 205 along the X and Y axes. Transfer path 215 may be the path along which the container 205 moves and may be defined by a body comprising, for example, PTFE material. The position of each stage (e.g., input / output (I / O) stage, capping stage, image capture stage, and image processing stage) performed along transfer path 215 may be based on the height or volume of the container 205. The determining device 100 may adjust the transfer speed of the container 205 along transfer path 215 depending on the process of the method including the respective stages.

[0081] The image capture stage (“third stage”) can be the process of capturing an image of the target sample. In the image capture stage, the determining device 100 may acquire an image of the target sample by at least two of the first camera 251, the second camera 253 and the third camera 255, as well as by at least one light source (e.g., bottom light source 257 and side light source 259) and other auxiliary devices (e.g., reflectors) for capture.

[0082] The first camera 251 may be a camera that captures a first image corresponding to the upper part of the target sample when the container is open. The first camera 251 may be referred to as an "upper capture camera". The second camera 253 may be a camera that captures a second image corresponding to the side of the target sample when the container is open. The second camera 253 may be referred to as a "side capture camera". The third camera 255 may be a camera that captures a third image corresponding to the lower part of the target sample. The third camera 255 may be referred to as a "lower capture camera". The third camera 255 can capture a third image corresponding to the lower part of the target sample regardless of whether the container is open.

[0083] At this time, the determining device 100 can use a pattern or patterned background image to capture an image of the target sample to obtain a structured light effect, thereby more clearly distinguishing the image of the target sample. The structured light can be a structured light source that projects a known pattern or patterned background image onto the captured image of the target sample. The patterned background image can include, but is not limited to, one or more of a white image, a grid pattern image, and / or a radial pattern image. The patterned background image can have various background patterns, various colors, and various brightness to facilitate the detection of the target to be detected (e.g., the opacity of the solution, the presence or absence of undissolved sample particles in the solution, and / or residues around the container, etc.). In addition, the patterned background image can have patterns designed in various forms according to, for example, the dissolution of the solution in which the current target sample is dissolved, and the properties of the solvent and solute (target sample) in which the target sample is dissolved.

[0084] The image processing stage (“fourth stage”) can be a process of automatically determining whether the target sample has dissolved from (multiple) captured images. In the image processing stage, the determining device 100 can perform image processing on the images acquired in the image capture stage using the image processing algorithm described above, thereby automatically determining whether the target sample has dissolved as “pass” or “fail” with higher accuracy. Here, the image processing algorithm can detect the size and number of particles in the undissolved solute from the captured image of the target sample. Furthermore, the image processing algorithm can measure the solubility, turbidity, and amount of remaining undissolved solute in the solution in real time.

[0085] Figure 3 This is a flowchart illustrating a method for operating the dissolution determination apparatus according to an embodiment. Reference will be made below. Figure 3 The operations described can be performed sequentially or non-sequentially. For example, the order of operations can be changed, and at least two operations can be performed in parallel.

[0086] refer to Figure 3 According to the embodiment, the determining device can automatically determine whether the target sample has been dissolved by operating 310 to 340.

[0087] In operation 310, the determining device can read the identification information of the container holding the target sample.

[0088] In operation 320, the determining device can control whether to open the container (e.g., remove the lid from the container) by moving the container containing the target sample to the capture position or moving the container containing the target sample toward the capture position. The determining device can transfer the container containing the target sample to the capture position along a transfer path. The determining device can adjust at least one of the height and horizontal position of the container. The determining device can control whether to open the container after adjusting at least one of the height and position of the container. For example, the determining device can control a capper to open the container.

[0089] In operation 330, the determining device can acquire an image of the target sample when the container is opened. The determining device can capture a first image corresponding to the upper part of the target sample when the container is opened. The determining device can capture a second image corresponding to the side of the target sample when the container is opened. The determining device can capture a third image corresponding to the lower part of the target sample.

[0090] In operation 340, the determining device can automatically determine whether the target sample has dissolved by analyzing an image of the target sample based on the container identification information read in operation 310. The determining device can extract the ROI from the image of the target sample. The following will refer to... Figure 5 A more detailed description of the method for detecting ROI using a deterministic device.

[0091] The determination device can determine whether a target sample has dissolved based on frequency-based filtering of the ROI. The determination device can also calculate the noise level corresponding to the ROI, for example, using image entropy.

[0092] The determination device can use image entropy to calculate the noise level of the ROI. Here, image entropy can be an indicator used to measure the complexity and uncertainty of an image. Higher entropy can mean that the image contains a lot of information and less noise. The determination device can evaluate the amount of information in the image based on the probability distribution of each pixel value in the image using entropy.

[0093] The method for calculating noise levels using image entropy is as follows. The determining device can calculate the entropy based on the probability distribution of each pixel value in the image, for example, by Equation 1 below.

[0094] [Equation 1]

[0095]

[0096] here, It can correspond to pixel values The probability of.

[0097] The determining device can calculate the entropy difference by comparing the entropy of a noise-free ideal image (e.g., an image in which the target sample is completely dissolved) with the entropy of a real image (e.g., an image of the target sample contained in a container). The determining device can use the entropy difference to estimate the noise level. References will follow below. Figure 6 A method for calculating noise levels using image entropy by a determining device is described in more detail.

[0098] The device can detect particles of undissolved solute included in the ROI by using frequency-based filtering with respect to the ROI. The following will refer to... Figure 4A method for detecting undissolved solute particles by a deterministic device is described in more detail.

[0099] The determination device can determine whether a target sample has been dissolved based on noise levels and undissolved solute particles.

[0100] According to an embodiment, the determining device can analyze the degree of dissolution of a target sample by extracting features from the Region of Interest (ROI) and applying the extracted features to a neural network trained based on an analysis algorithm. The determining device can generate a control signal based on the analysis results. The analysis algorithm can analyze at least one of the following: whether the target sample is completely dissolved, the opacity of the target sample, undissolved solute particles in the target sample, and residues around the container.

[0101] Alternatively, the determining device can determine the dissolution conditions corresponding to the target sample based on the container's identification information. The determining device can generate a control signal to accelerate the dissolution of the target sample based on the dissolution conditions, depending on whether the target sample has been completely dissolved. When it is determined that the target sample has not been completely dissolved, the determining device can generate a control signal to accelerate the dissolution of the target sample based on the dissolution conditions. The dissolution conditions may include at least one of the following: the type of solvent used for dissolving the target sample; the amount of solvent; the type of catalyst used for dissolving the target sample; the amount of catalyst, temperature, humidity, and / or pressure; and the number of times the container is agitated, but are not limited thereto.

[0102] Figure 4 This is a flowchart illustrating a method for operating the dissolution determination apparatus according to an embodiment. (See reference...) Figure 4 According to the embodiment, the determining device can automatically determine whether the target sample has been dissolved by operating 410 to 450.

[0103] In operation 410, the determining device can move (multiple) cameras to a specific location. Here, "specific location" can be, for example, the location for precisely capturing an image of at least one container holding a solution in which the target sample is dissolved. The determining device can move and fix the positions of the cameras (e.g., a first camera, a second camera, and / or a third camera) included in the image acquisition module to this specific location via the aforementioned drive module.

[0104] In operation 420, the determining device can control the light source to obtain at least one image (e.g., an image of the target sample).

[0105] In operation 430, the determining device can detect a ROI within at least one image obtained in operation 420. The determining device can detect a container by its outline within at least one image, and detect the ROI inside the container. Alternatively, the determining device can detect the ROI from at least one image using the geometric properties, color information, etc., of the solvent and / or solute (target sample). Reference will be made below. Figure 5 A more detailed description of the method for detecting ROI using a deterministic device.

[0106] In operation 440, the determining device may perform frequency-based filtering with respect to the ROI detected in operation 430. Frequency-based filtering can be used to remove noise from specific parts of the image or to highlight specific frequency components. The determining device may, for example, transform the ROI to the frequency domain using a Discrete Fourier Transform (DFT). The determining device may then transform the frequency domain back to the spatial domain by applying various filters, such as a low-pass filter (LPF) or a high-pass filter (HPF), to the ROI. The determining device may then remove noise from specific parts of the image or highlight specific frequency components by applying the filtering results to the ROI.

[0107] In the frequency-based filtering process of operation 440, the determination device can calculate the noise level of the ROI through operation 441. (See below for reference.) Figure 6 The method for calculating the noise level of ROI by a determining device is described in more detail.

[0108] Additionally, during the frequency-based filtering process in operation 440, the determining device can detect undissolved solute particles via operation 443. The determining device can use at least one of a bandpass filter, an adaptive threshold, and a nonlinear filter to detect undissolved solute particles. Alternatively, the determining device can detect undissolved solute particles in a solution containing the target sample by applying a fine particle tracking algorithm.

[0109] The determination device can use an adaptive thresholding scheme to detect particles of undissolved solute. This adaptive thresholding scheme is used to obtain a clearer image by dividing the image (ROI) into multiple regions, calculating the average or Gaussian value of the region based on the surrounding pixel values, and specifying a threshold for each region.

[0110] In operation 450, the determining device can determine whether the target sample has been dissolved based on the noise level calculated in operation 441 and the undissolved solute particles detected in operation 443.

[0111] In operation 460, the determining device can output an indication of whether the target sample has been dissolved based on the determination in operation 450. For example, when it is determined in operation 450 that the target sample is in a state such as Figure 7 The determining device may output "failure" when the target sample is in an undissolved state as shown in schematic diagram 710, i.e., a supersaturated state in which solute particles are present in the solution. Alternatively, when the target sample is determined to be in such a state in operation 450... Figure 7 When the device is in a completely dissolved state as shown in the schematic diagram 720, it can output "pass".

[0112] Figure 5 This is a schematic diagram illustrating a method for extracting a Region of Interest (ROI) from an image of a target sample according to an embodiment. (Reference) Figure 5 Various images 510, 520, 530 and 540 of the target sample according to the embodiments are shown.

[0113] The determining device can detect the Region of Interest (ROI) within the container by detecting the container outlines 515, 525, 535, and 545 within the images 510, 520, 530, and 540 of the target sample. In other words, the determining device can detect the region within the container outlines 515, 525, 535, and 545 as the ROI.

[0114] More specifically, since there may be out-of-focus images in the images 510, 520, 530 and 540 of the target sample, the determining device can blur the images 510, 520, 530 and 540 of the target sample so that the images are in the same frequency band.

[0115] The determining device can apply HPF to images of target samples that are in the same frequency band and generate binarized images corresponding to images 510, 520, 530, and 540 of the target samples by applying appropriate thresholds. The determining device can detect ROIs by applying the Hough circle detection method to the binarized images to detect the contours 515, 525, 535, and 545 of the container.

[0116] The Hough circle detection method corresponds to an algorithm used to find circles in an image. The Hough circle detection method uses the Hough transform to detect the center and radius of the circle. The determination device can detect edges in a binarized image, for example, using a Canny edge detector. The determination device can apply the Hough transform to detect circles in the image of the detected edges. The Hough transform method extracts the circle by selecting a two-dimensional histogram of the center points (a, b) of the circle from the detected edges in the image using a gradient method, and increasing the distance of all points in the accumulation plane from minimum to maximum distance along the gradient line segment for all points. Here, the accumulation plane can be a three-dimensional accumulation plane including the center point x of the circle, the center point y of the circle, and the radius r of the circle. The determination device can display the detected circles (e.g., the outlines 515, 525, 535, and 545 of the container) on the original image and detect the regions within the outlines 515, 525, 535, and 545 of the container as ROIs.

[0117] Alternatively, the determination device can use the geometric properties, color information, etc. of the solvent and / or solute (target sample) to detect the ROI from the image.

[0118] Figure 6 This is a schematic diagram illustrating a method for calculating the noise level corresponding to the ROI according to an embodiment.

[0119] Since the cameras that have captured images (images of the target sample) may have different focal points, the determining device can perform blurring (e.g., Gaussian blur) to bring the ROI of each image into the same frequency band. When blurring is not performed, the focused and out-of-focus images may not be processed the same way.

[0120] The determining device can detect noise in an image by applying an edge filter to a blurred Region of Interest (ROI). Noise in the image may correspond to particles of floating matter included within the ROI and / or undissolved solutes included within the ROI. According to an embodiment, the determining device may perform normalization and / or discretization after applying the edge filter to the blurred ROI.

[0121] The determining device can perform a primary mask for information about the passage. Figure 5 The detected container outlines 515, 525, 535, and 545 are used to mask the image of the target sample. The determining device can accurately detect and / or determine the solute particles inside the container by, for example, masking the exterior of the container outlines 515, 525, 535, and 545 to eliminate the influence of reflections on the image from the container walls.

[0122] In a comparative embodiment, during the binarization operation, the determining device may detect particles of the internal solute based on the edges, and in this process, the container contour may have an excessive influence, which may lead to false detections. In embodiments of this disclosure, false detections can be addressed by using a mask, and the masking operation can affect the overall distribution of pixels (such as binarization) to improve detection performance. For example, in areas where there is heavy exposure due to uneven exposure, problems such as local saturation where edges also become enlarged may occur, and therefore, in the comparative embodiment, it may be difficult to extract the desired information using a simple threshold.

[0123] In embodiments of this disclosure, an adaptive threshold that takes background removal into account can be used to detect noise in an image. The adaptive threshold can be a technique that performs binarization by dynamically determining a threshold for each portion of the image. This can effectively remove the background or detect objects, even in images with uneven lighting.

[0124] The determining device can determine whether a target sample has been dissolved by detecting noise in an image.

[0125] Here, schematic diagrams 610, 620, 630, and 640 can be instantaneous images output as a result of masking the actual measured camera images. Schematic diagrams 610 and 640 can represent images when undissolved material is present, and schematic diagrams 620 and 630 can represent images when the solute is completely dissolved.

[0126] In graph 650, values ​​on the y-axis can be assigned based on the dissolution state (e.g., pass or fail) determined by the determining device. In graph 650, pass data (e.g., data points corresponding to images where the target sample is determined to be completely dissolved) can have x-axis values ​​between 1.95 and 2.60, and fail data (e.g., data points corresponding to images where the target sample is determined to be incompletely dissolved) can have x-axis values ​​between 2.33 and 2.87. The values ​​on the x-axis can represent the entropy values ​​of the image. Graph 650 shows the distribution of image entropy for each of the dissolved (pass) and indissolved (fail) cases. In graph 650, the values ​​of data points with respect to the y-axis can be represented as 0 or 1 to indicate whether the target sample in the corresponding image is determined to be completely dissolved. The determining device can use the image entropy shown in graph 650 as one of the criteria for determining whether the target sample has been dissolved.

[0127] Figure 7 This is a schematic diagram illustrating a method for determining whether a target sample has been dissolved according to an embodiment.

[0128] The determining device according to the embodiment can automatically determine whether a target sample has been dissolved based on image analysis results of the target sample. The determining device can utilize the overall distribution of passes and failures based on the aforementioned image entropy as a feature for determining whether the target sample has been dissolved.

[0129] The determining device can apply a median filter to an image of a target sample. The determining device can remove dust from the image (multiple) of a camera and / or a container by applying the median filter to the image of the target sample. The median filter can be a non-linear digital filter that can be used to remove noise from an image or signal. The median filter can remove noise using the median of the surrounding pixel values. The determining device can set a predetermined region (window) around the pixel to be filtered. The determining device can set the window to a size of, for example, 3×3 or 5×5. The determining device can align the pixel values ​​within the window and then select their median. The selected median can be the new pixel value for the corresponding pixel. The determining device can remove salt-and-pepper noise from the image of the target sample by applying the median filter. Salt-and-pepper noise can correspond to noise in the form of randomly generated white and black dots in an image. Salt-and-pepper noise can be primarily generated when the pixel value of the image suddenly changes to 0 (black) or 255 (white).

[0130] As referenced above Figure 6 As mentioned above, edges may appear large in corresponding portions due to the influence of edge detection and the primary masking of the detected container contour. In an embodiment, secondary masking of the ROI can be performed using a mask with a radius slightly smaller than the radius of the ROI to remove the effect of edges appearing large in the image. The determining device can count the number of particles remaining in the image as a result of the secondary masking and use the number of particles as a feature for determining whether the target sample has been dissolved. The determining device can determine whether the target sample has been dissolved based on a comparison between the number of particles and a threshold.

[0131] For example, when the number of remaining particles in the image (such as in schematic diagram 710 and graph 730) exceeds a set threshold (e.g., 5), the determining device can determine the dissolution state as an undissolved state, i.e., where the solute particles exist in a supersaturated state in the solution, and output "failure".

[0132] For example, the threshold can be a value greater than the number of solute particles included in a solution in a fully dissolved state, in which solute particles are almost non-existent in the solution due to the high solubility of the solution in which the target sample is dissolved, and can be a value less than or equal to the number of solute particles included in a crystalline (or supersaturated) state, in which the solute extracted from the image is well dissolved in the solvent but no longer exists as solute particles at a predetermined concentration or higher, but is not limited thereto.

[0133] The determination device can determine whether the type of analytical result (failure) is "Failure #1" corresponding to a turbid state or "Failure #2" corresponding to a "supersaturated state". In addition to the number of solute particles, the determination device can determine whether the target sample has been dissolved by considering the transparency (or opacity) of the solution shown in the ROI and / or the presence of residue around the container.

[0134] For example, when the number of solute particles greatly exceeds a threshold and the opacity of the solution displayed in the ROI is high, the determining device can determine that the type of the analytical result (failure) is "Failure #1" corresponding to "turbidity". Conversely, when the number of solute particles is slightly greater than or equal to the threshold and the transparency of the solution displayed in the ROI is high, the determining device can determine that the type of the analytical result (failure) is "Failure #2" corresponding to "supersaturation". For example, the determining device can compare the number of solute particles with an additional threshold that is greater than the threshold to determine whether the type of the analytical result (failure) is "Failure #1" or "Failure #2". For example, when the number of solute particles is equal to or greater than the additional threshold, the determining device can determine that the type of the analytical result (failure) is "Failure #1", and when the number of solute particles is less than the additional threshold, the determining device can determine that the type of the analytical result (failure) is "Failure #2".

[0135] When the analysis result (failure) is determined to be turbid ("Failure #1"), the determining device can generate a control signal to adjust the temperature, agitation speed, or reaction time to increase the solubility of the target sample, and send this control signal to the drive module, causing the determining device to control the adjustment of the temperature, agitation speed, or reaction time. The determining device can then obtain an image after adjusting the temperature, agitation speed, or increasing the reaction time according to the control signal.

[0136] When the analysis result (failure) is determined to be a supersaturated state ("Failure #2"), the determining device can generate a control signal for adding solvent and send this control signal to the drive module, causing the determining device to initiate the addition of solvent. The determining device can then obtain an image after adding solvent according to the control signal.

[0137] For example, when the number of remaining particles in the image (such as in schematic diagram 710 and graph 730) is less than or equal to a set threshold (e.g., 5), the determining device can determine the dissolution state as a completely dissolved state and output "pass".

[0138] Figure 8 This is a flowchart illustrating a method for operating the dissolution determination apparatus according to an embodiment. (See reference...) Figure 8 According to the embodiment, the determining device can perform the operation of determining whether the target sample has been dissolved by operating 801 to 843.

[0139] In operation 801, the determining device can move the target container to an initial position. At this point, the determining device can move the target container to the initial position using a general-purpose robot 5 (UR5). UR5 can be a robotic arm (e.g., a robotic hand or gripper) for transferring the target container. UR5 can be used to transfer the target container containing the reagent from a dispensing apparatus or agitation device to a dissolution measurement device. After the dissolution measurement is completed, UR5 can transfer the target container to another device for further processing.

[0140] In operation 803, it is determined that the device can move the target container to the location of the QR reader.

[0141] In operation 805, the determining device can identify information about the target container while verifying its existence using a QR reader.

[0142] In operation 807, the determining device can move the target container to the position of the capper using a gantry robot. At this point, after moving the target container to the position of the capper, the gantry robot can move the target container to another position without obstructing image capture.

[0143] In operation 809, the determining device can adjust the height of the target container via a height adjusting motor (e.g., an actuator). The determining device can adjust the height of the target container such that the position of the lid of the target container reaches the position of the capping device.

[0144] In operation 811, the determining device can perform unsealing of the target container via the capper. For example, the determining device can control the capper to remove the cap from the target container.

[0145] In operation 813, the determining device can return the target container from its adjusted height in operation 809 to its original height by means of a height adjusting motor (e.g., an actuator).

[0146] In operation 815, the determination device can move the target container to a platform provided with bottom and side light sources by means of a gantry robot.

[0147] In operation 817, the determining device can automatically set the positions of (multiple) cameras (e.g., a first camera, a second camera, and / or a third camera) according to the volume of the target container. At this time, the camera positions can correspond to locations where unobstructed imaging of the upper part, the side, and the lower part of the target sample contained within the target container is possible.

[0148] In operation 819, the determining device can automatically set the illuminance of the light source (e.g., side light source and bottom light source) according to the volume of the target container.

[0149] In operation 821, the determining device can automatically set measurement conditions based on the volume of the target container. Measurement conditions can be those used for image capture, such as white balance, sharpness, and exposure time. White balance is primarily used in photography or video recording to adjust the color temperature to express natural colors. Sharpness refers to the clarity and detail of an image in a photograph or video, and higher sharpness means a clearer image with better detail.

[0150] In operation 825, the determining device can capture a third image of the lower part of the target container according to the setting conditions of operations 817 to 821. The determining device can capture the third image when the third camera is facing the lower part (e.g., the bottom) of the target container, such that the third image is the lower part of the target sample. For example, the third camera can capture the third image when facing upwards toward the bottom of the target container.

[0151] In operation 829, the determining device can capture a second image facing the side of the target container according to the setting conditions of operations 817 to 821. The determining device can capture the second image when the second camera faces the side of the target container, such that the second image is the side of the target sample. For example, the second camera can capture the second image when it is sideways facing the side of the target container.

[0152] In operation 831, the determining device can determine whether the target sample has been dissolved based on the third image captured in operation 825 and the second image captured in operation 829 and output an indication of the determination (e.g., a dissolution result value (e.g., pass or fail)).

[0153] In operation 833, the determination device can move the target container to the capping stage, in which the capper is placed, using a gantry robot.

[0154] In operation 835, the determining device can adjust the height of the target container via a height adjusting motor (e.g., an actuator). The determining device can raise the height of the target container so that the position of the lid of the target container reaches the position of the capper, thereby performing the capping of the target container.

[0155] In operation 837, the determining device can perform capping of the target container via a capping device. For example, the determining device can control the capping device to attach the cap to the target container.

[0156] In operation 839, the determining device can return the target container from its adjusted height in operation 835 to its original height (e.g., the height of the target container immediately preceding operation 835) by controlling a height adjusting motor (e.g., an actuator).

[0157] In operation 841, the determining device can transfer the target container to the initial position using a gantry robot.

[0158] In operation 843, the determination device can remove the target container from its initial position via UR5.

[0159] The embodiments described herein can be implemented using hardware components, software components, and / or combinations thereof. The processing device can be implemented using one or more general-purpose or special-purpose computers, such as, for example, processors, controllers, arithmetic logic units (ALUs), digital signal processors (DSPs), microcomputers, field-programmable gate arrays (FPGAs), programmable logic units (PLUs), microprocessors, or any other device capable of responding to and executing instructions in a defined manner. The processing device may run an operating system (OS) and one or more software applications running on the OS. The processing device may also access, store, manipulate, process, and generate data in response to the execution of the software. For simplicity, the description of the processing device is used as the singular; however, those skilled in the art will appreciate that the processing device may include multiple processing elements and / or various types of processing elements. For example, the processing device may include multiple processors, or a single processor and a single controller. Furthermore, different processing configurations are possible, such as parallel processors.

[0160] Software can include computer programs, code segments, instructions, or some combination thereof, to independently or consistently instruct or configure a processing device to operate as needed. Software and data can be permanently or temporarily embodied in any type of machine, component, physical or virtual device, or computer storage medium or device capable of providing instructions or data to or being interpreted by the processing device. Software can also be distributed across network-coupled computer systems, enabling it to be stored and executed in a distributed manner. Software and data can be stored on one or more non-transitory computer-readable recording media.

[0161] The methods according to the above embodiments can be recorded in a non-transitory computer-readable medium including program instructions for implementing the various operations of the above embodiments. The medium may also include data files, data structures, etc., alone or in combination with the program instructions. The program instructions recorded on the medium may be those specifically designed and constructed for the purposes of the embodiments, or they may be of a type well known and available to those skilled in the art of computer software. Examples of non-transitory computer-readable media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical media such as CD-ROMs, DVDs, and / or Blu-ray discs; magneto-optical media such as optical discs; and hardware devices specifically configured to store and execute program instructions, such as read-only memory (ROM), random access memory (RAM), flash memory (e.g., USB flash drives, memory cards, memory sticks, etc.). Examples of program instructions include machine code generated by a compiler and files containing higher-level code that can be executed by a computer using an interpreter.

[0162] The aforementioned hardware device can be configured to act as one or more software modules to perform the operations of the above embodiments, and vice versa.

[0163] While non-limiting exemplary embodiments have been described with reference to the accompanying drawings, those skilled in the art can apply various technical modifications and variations based on these embodiments. For example, suitable results may be achieved if the described techniques are performed in a different order and / or if components in the described system, architecture, device, or circuit are combined in different ways and / or replaced or supplemented by other components or their equivalents. Therefore, other implementations, other embodiments, and equivalents including the above modifications and variations are included within the spirit and scope of this disclosure.

Claims

1. A dissolution determination apparatus, comprising: The reader is configured to read the identification information of the container holding the target sample; A drive module includes at least one body, the at least one body of the drive module being configured to open the container and move the container to a capture position; An image acquisition module is configured to acquire at least one image of the target sample at the capture location when the container is opened; as well as At least one processor is configured to automatically determine whether the target sample has been dissolved by analyzing at least one image of the target sample based on the identification information of the container.

2. The dissolution determination apparatus according to claim 1, wherein, The reader is further configured to sense the position of the container.

3. The dissolution determining apparatus according to claim 1, wherein, The at least one body of the drive module includes at least one of the following: A gripper is configured to grip the container; The robot is configured to transfer the container to the capture location along a transfer path; An actuator is configured to adjust at least one of the height and horizontal position of the container; as well as A capping device is configured to open the container.

4. The dissolution determination apparatus according to claim 1, wherein, The image acquisition module includes at least one of the following: A first camera is configured to capture a first image of the upper part of the target sample when the target sample is in the container and the container is opened; A second camera is configured to capture a second image of the side of the target sample when the target sample is in the container and the container is opened; as well as A third camera is configured to capture a third image of the lower part of the target sample when the target sample is in the container, and The at least one processor is configured to automatically determine whether the target sample has been dissolved by analyzing at least one of the first image, the second image, and the third image of the target sample based on the identification information of the container.

5. The dissolution determining apparatus according to claim 1, wherein, The at least one processor is further configured to: Extract the region of interest (ROI) from at least one image of the target sample; The noise level corresponding to the ROI is calculated by frequency-based filtering of the ROI; Detection includes particles containing undissolved solutes in the ROI; as well as Whether the target sample has been dissolved is determined based on the noise level and the particles of the undissolved solute.

6. The dissolution determining apparatus according to claim 5, wherein, The at least one processor is further configured to detect the particles of the undissolved solute using at least one of a bandpass filter, an adaptive threshold, and a nonlinear filter.

7. The dissolution determination apparatus according to claim 5, wherein, The at least one processor is further configured to: Extract features from the ROI; The analytical results are obtained by analyzing the degree of solubility of the target sample by applying the features to a neural network trained based on the analytical algorithm; as well as Control signals are generated based on the analysis results.

8. The dissolution determination apparatus according to claim 7, wherein, The analysis algorithm is configured to analyze at least one of the following: whether the target sample is completely dissolved, the opacity of the target sample, the particles of the undissolved solute in the target sample, and the residue around the container.

9. The dissolution determination apparatus according to claim 1, wherein, The at least one processor is further configured to: The dissolution conditions corresponding to the target sample are determined based on the identification information of the container; as well as Based on the determination that the target sample has not been completely dissolved, a control signal is generated to accelerate the dissolution of the target sample according to the dissolution conditions.

10. The dissolution determination apparatus according to claim 9, wherein, The dissolution conditions include at least one of the following: the type of solvent used for dissolving the target sample; the amount of the solvent; the type of catalyst used for dissolving the target sample; the amount, temperature, humidity, or pressure of the catalyst; and the number of times the container is agitated.

11. A method performed by a dissolution determination apparatus, the method comprising: Read the identification information of the container holding the target sample; Open the container; Move the container to the capture location; When the container contains the target sample and the container is open, at least one image of the target sample is obtained at the capture location; as well as Whether the target sample has been dissolved is automatically determined by analyzing at least one image of the target sample based on the identification information of the container.

12. The method according to claim 11, wherein, Opening the container includes: Adjusting at least one of the height and position of the container; and The container is opened after at least one of the height and position of the container is adjusted.

13. The method according to claim 11, wherein, Obtaining the at least one image of the target sample includes: A first image of the upper part of the target sample is captured when the container contains the target sample and is opened; A second image of the side of the target sample is captured when the container contains the target sample and is opened; and A third image of the lower part of the target sample is captured when the container contains the target sample.

14. The method according to claim 11, wherein, Automatically determining whether the target sample has been dissolved includes: Extracting the region of interest (ROI) from at least one image of the target sample; and Whether the target sample is dissolved is determined based on frequency-based filtering of the ROI.

15. The method according to claim 14, wherein, Determining whether the target sample has been dissolved based on the frequency-based filtering includes: Image entropy is used to calculate the noise level corresponding to the ROI; Particles of undissolved solute included in the ROI are detected by frequency-based filtering of the ROI; and Whether the target sample has been dissolved is determined based on the noise level and the particles of the undissolved solute.

16. The method according to claim 15, wherein, Detecting the particles of the undissolved solute includes: The particles of the undissolved solute are detected using at least one of a bandpass filter, an adaptive threshold, and a nonlinear filter.

17. The method of claim 14, wherein, Automatically determining whether the target sample has been dissolved further includes: Extract features from the ROI; The analytical results are obtained by analyzing the solubility of the target sample by applying the features to a neural network trained based on an analytical algorithm; and Control signals are generated based on the analysis results.

18. The method according to claim 17, wherein, The analysis algorithm analyzes at least one of the following: whether the target sample is completely dissolved, the opacity of the target sample, undissolved solute particles in the target sample, and residues around the container.

19. The method according to claim 11, wherein, Automatically determining whether the target sample has been dissolved includes: The dissolution conditions corresponding to the target sample are determined based on the identification information of the container; and Based on the determination that the target sample has not been completely dissolved, a control signal is generated to accelerate the dissolution of the target sample according to the dissolution conditions.

20. The method according to claim 19, wherein, The dissolution conditions include at least one of the following: the type of solvent used for dissolving the target sample; the amount of the solvent; the type of catalyst used for dissolving the target sample; the amount, temperature, humidity, or pressure of the catalyst; and the number of times the container is agitated.

Images