A full-automatic photographing system for cordyceps

Abstract

The application discloses a kind of cordyceps sinensis full-automatic photographing system, belong to machine vision and digital technology field of traditional Chinese medicinal materials.The system includes closed acquisition cabin, precision mechanical transmission module, multi-modal perception module and edge computing control unit.Core uses "bottom MCU+upper computer NPU" dual-core architecture, through closed dark box design to isolate extraneous stray light, utilize precision rotary table to realize sample 360° dead angle shooting.System synchronously obtains high-definition image, accurate weight and GPS geographic information in one acquisition cycle, and carries out encryption binding storage.The application realizes the full-process standardization and automation of cordyceps sinensis image acquisition, solves the problem of uneven illumination, large environmental interference and data fragmentation in field collection, generates multi-dimensional fusion data set conforming to AI training gold standard, and has the advantages of strong adaptability to plateau environment, high data integrity and good portability.

Description

Technical Field

[0001] This invention relates to the fields of machine vision, digital technology of traditional Chinese medicine and embedded systems, specifically a fully automatic Cordyceps sinensis photography system that is adapted to the wild plateau environment and has multi-dimensional information binding function. Background Technology

[0002] Cordyceps sinensis is a precious traditional Chinese medicine, considered one of the three major traditional tonics alongside ginseng and deer antler, and is hailed as "the world's number one herb," ​​"the sacred herb of the East," "the king of medicines," and "the gold of the snowy plateau." It is formed by the infection of Cordyceps sinensis fungus, which parasitizes the larvae of moths in the family Hepialidae. It is mainly distributed in alpine meadow areas at altitudes of 3500-5000 meters, such as Qinghai, Tibet, Sichuan, and Yunnan. Cordyceps sinensis contains effective components such as polysaccharides, cordyceps, adenosine, mannitol, sterols, mannan, and nucleosides, possessing therapeutic properties including anti-aging, anti-cancer / anti-tumor, anti-inflammatory, immune-stimulating, and antioxidant effects. Among these, "Bailing Capsules," made primarily from Cordyceps sinensis, have been widely used to treat diseases such as asthma and tuberculosis.

[0003] Cordyceps sinensis, a rare medicinal fungus, relies heavily on morphological characteristics for its efficacy and quality assessment. Currently, image acquisition within the industry faces the following challenges: 1) Low standardization: Existing technologies mostly rely on manual handheld shooting or simple fixed brackets. Shooting is greatly affected by the photographer's experience, lighting angle, and shooting distance, resulting in uneven lighting and angular deviations in image data, which cannot meet the requirements of subsequent AI 3D reconstruction and deep learning for high-quality datasets (gold standard).

[0004] 2) Significant environmental interference: The main production areas of Cordyceps sinensis are located in high-altitude areas such as the Qinghai-Tibet Plateau (e.g., Nagqu and Yushu), where natural light changes drastically and there is a lot of wind and sand. Shooting in open environments is easily affected by external stray light, weather, and vibration, resulting in unstable image quality.

[0005] 3) Data fragmentation: Traditional image acquisition and metadata such as sample weight, geographical location (GPS), and acquisition time are often recorded separately, requiring manual entry later, which can easily cause data misalignment and lack unique identifiers.

[0006] Therefore, there is an urgent need for a fully automated Cordyceps sinensis photography system that can achieve closed-loop, standardized, and multi-dimensional data fusion. Summary of the Invention

[0007] To address the aforementioned problems, this invention provides a fully automatic photographing system for Cordyceps sinensis.

[0008] Specifically, the fully automatic Cordyceps sinensis photography system includes a closed collection chamber, a precision mechanical transmission module, a multimodal sensing module, an edge computing control unit, and a power management system. Specifically, the enclosed collection chamber is equipped with a shooting space, and the precision mechanical transmission module is located in the shooting space to fix and rotate the Cordyceps sinensis sample; Specifically, the multimodal perception module includes a high-definition camera and a supplementary lighting unit; the edge computing control unit includes a bottom-level MCU and a host computer NPU, used to control mechanical transmission and image acquisition.

[0009] Specifically, the enclosed collection chamber adopts an integrated handle design, with overall dimensions of 300mm (length) × 300mm (width) × 300mm (height) and a total weight of less than 6kg.

[0010] Specifically, the multimodal sensing module also includes a high-precision weighing sensor and a GPS positioning module. The edge computing control unit is configured to simultaneously acquire image, weight and location data within a single acquisition cycle, and to encrypt and bind the data for storage.

[0011] Specifically, the precision mechanical transmission module includes a stepper motor, a coupling, and a sample fixing fixture; the sample fixing fixture is a detachable clamp, and the stepper motor drives the turntable to rotate 360° with a positioning accuracy of 0.1°.

[0012] Specifically, the edge computing control unit adopts a dual-core main control architecture of "MCU + NPU": Specifically, the underlying MCU uses an STM32F4 series chip, which is responsible for real-time control of the stepper motor, supplementary light, and reading of weighing data; Specifically, the host computer NPU uses an NVIDIA edge computing module (Jetson Nano) to handle image processing, human-computer interaction, 5G data encryption upload, and local storage management.

[0013] Specifically, the high-precision weight sensor is integrated below the sample placement platform, with a weighing accuracy of no less than 0.01g; the system also includes a GPS positioning module for acquiring the latitude, longitude, and altitude information of the collection point.

[0014] Specifically, the power management system integrates a 24V lithium battery pack, and with intelligent power consumption strategy, the battery life is ≥12 hours when fully charged; the data transmission supports 5G / WIFI communication and has the function of resuming interrupted transmission, and automatically caches data to the local solid-state drive when the network signal is weak.

[0015] Specifically, the system has built-in automated control software. After the user scans the code to enter the sample ID, the system automatically controls the rotation, shooting (collecting 4-8 multi-angle images each time), weighing, and encrypting and packaging the image data with weight, GPS, and time metadata for synchronization.

[0016] A method for collecting data on Cordyceps sinensis includes the following steps: S1 Enter the sample ID; S2 controls the activation of the supplementary lighting unit and drives the sample to rotate to a preset angle; S3 triggers the high-definition camera to acquire images; S4 reads data from the weighing sensor and GPS module, and binds and stores them with image data.

[0017] Specifically, in step S4, if the network signal strength is lower than a preset threshold, the system automatically caches the data locally and automatically performs incremental synchronization after the signal is restored.

[0018] Specifically, the acquired multi-angle images are reconstructed in three dimensions using the 3D Gaussian Splatting algorithm or the SfM algorithm to generate a digital three-dimensional model of Cordyceps sinensis.

[0019] It has the following beneficial effects: (1) Full-process standardization: Through a closed dark box and automated robotic arm, the interference of natural light and human operation is eliminated, ensuring that the lighting, angle and distance of each Cordyceps sinensis image are consistent, which meets the gold standard of AI training.

[0020] (2) Strong adaptability to plateau environment: The fully enclosed design and industrial-grade components enable it to adapt to high altitude, low air pressure, low temperature (-10℃ to 40℃) and sandstorm environment; the integrated portable design solves the problem of equipment transportation in high altitude areas.

[0021] (3) Multidimensional data fusion: In a single acquisition process, high-definition images, three-dimensional pose information (through multi-angle stitching), precise weight and geographic information are acquired simultaneously to construct a complete digital archive of the sample and avoid data misalignment.

[0022] (4) High efficiency and intelligence: The dual-core architecture has a clear division of labor. The bottom layer ensures real-time control, while the upper computer is responsible for high-performance computing, realizing unattended operation from data acquisition to uploading. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the structure of a fully automatic Cordyceps sinensis photography system according to the present invention; Figure 2 This is an electrical connection block diagram of a fully automatic Cordyceps sinensis photography system according to the present invention; Figure 3 This is a control flowchart of a fully automatic Cordyceps sinensis photography system according to the present invention; Reference numerals: 1-Enclosed acquisition chamber, 2-Edge computing control unit, 3-Power management system, 4-High-definition camera, 5-Precision mechanical transmission module, 6-Supplemental lighting unit, 7-Multimodal sensing module. Detailed Implementation

[0025] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0026] The following detailed description of the implementation method of the present invention is in conjunction with the accompanying drawings. The description is only a partial embodiment and not all embodiments. For clarity, representations and descriptions unrelated to the present invention are omitted in the drawings and description.

[0027] To provide a clearer understanding of the technical features, objectives, and beneficial effects of this invention, the following detailed description of the technical solution is provided. Obviously, the described embodiments are only a portion of the embodiments of this invention, not all of them, and should not be construed as limiting the scope of implementation of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the protection scope of this invention.

[0028] Example 1 This embodiment provides a fully automated Cordyceps sinensis photography system, aiming to solve the problems of bulky field collection equipment, significant interference from ambient light in imaging, limited data dimensions, and low processing efficiency in existing technologies. The system mainly includes: a closed collection chamber 1, a precision mechanical transmission module 5, a multimodal sensing module 7, an edge computing control unit 2, and a power management system 3.

[0029] In this embodiment, to adapt to the needs of high-altitude field operations, the system has undergone targeted optimization in terms of structure and hardware selection: In this embodiment, a portable, fully enclosed optical dark box design is adopted. The enclosed collection chamber 1 adopts an integrated handle design, and the overall size is strictly controlled within 300mm in length × 300mm in width × 300mm in height. The weight of the whole machine is less than 6kg, which makes it easy for a single person to carry to collection points at an altitude of over 4500 meters.

[0030] In this embodiment, the cabin frame is constructed using industrial-grade high-strength 6061 aluminum alloy profiles with anodized surface treatment; the doors, windows, and outer shell are made of black acrylic material (light transmittance <1%).

[0031] This fully enclosed design effectively isolates external stray light (400nm-700nm band) and wind and sand, creating a stable internal imaging environment.

[0032] In this embodiment, the drive system is equipped with a precision rotary table, which is driven by the underlying MCU through pulse signals to drive 42 stepper motors (256 microstepping drive) to achieve 360° infinite rotation with a positioning accuracy of 0.1°.

[0033] In this embodiment, considering the small size and complex texture of Cordyceps sinensis fruiting bodies, a specially made stainless steel needle with a diameter of 0.2 mm is used for clamping. This steel needle is directly inserted into the head or tail of the Cordyceps sinensis for fixation. Compared with traditional clamps, its obstruction area is negligible. Combined with a rotating stage, it can ensure complete imaging of the texture of the Cordyceps sinensis fruiting body and head without blind spots.

[0034] In this embodiment, the multimodal perception module 7 is equipped with an imaging unit featuring a high-definition camera module (such as IMX477), which supports 12-megapixel and 4K resolution image acquisition, and a diffuse reflector is added in front of the lens to eliminate reflections.

[0035] In this embodiment, the multimodal sensing module 7 integrates an Hx711 weighing module (range 0-30g, accuracy 0.01g) at the bottom and a NEO-6M GPS module on the side for acquiring latitude, longitude, and altitude information. The cabin is also equipped with a high color rendering index (Ra>90) LED supplementary lighting panel that supports stepless dimming.

[0036] In this embodiment, the core innovation lies in the adoption of a dual-core main control architecture of "bottom-level MCU + upper-level NPU", which achieves precise timing synchronization of software and hardware: In this embodiment, the underlying MCU (real-time control layer) uses an STM32F4 series chip (such as STM32F407). It is responsible for the real-time control of the underlying hardware. Specifically, this includes adjusting the brightness of the auxiliary light by outputting PWM signals through the GPIO port; receiving commands from the host computer and sending pulses to drive the stepper motor to rotate precisely; and reading raw data from the weighing sensor and GPS module through the SPI / I2C interface.

[0037] In this embodiment, the host computer NPU (edge ​​computing layer) uses an NVIDIA Jetson Nano edge computing module. It runs a Linux operating system and acquisition software developed based on Python / QT. It is responsible for image processing, human-computer interface display, 5G / WIFI encrypted data upload, and local SSD storage management.

[0038] In this embodiment, after the user enters the sample ID via touch screen or by scanning a code, the host computer issues a "collection command".

[0039] In this embodiment, the MCU receives a command and controls the motor to rotate 45° (preset at 8 angle points).

[0040] In this embodiment, after the motor stops, the MCU sends a "hardware ready signal" to the host computer.

[0041] In this embodiment, the host computer triggers the camera shutter to take a picture the instant it receives the signal, and simultaneously requests the current weight and GPS data from the slave computer.

[0042] In this embodiment, the mechanism ensures that the metadata (location, weight, time) of each image corresponds strictly to the image itself.

[0043] In this embodiment, the multimodal perception module not only acquires two-dimensional images, but also constructs multidimensional data packets and performs advanced edge processing: In this embodiment, multi-dimensional data binding is performed. Within a single acquisition cycle, the system generates a standard data package containing the following fields: Image data includes 8 high-resolution JPG images (named ID_angle.jpg) from different angles (one image per 45°). Attribute data includes weight (0.01g accuracy), geographic location (latitude and longitude + altitude), environmental parameters (cabin illumination, temperature), and timestamp.

[0044] In this embodiment, the above data is encrypted and packaged into a JSON file and stored together with the image.

[0045] In this embodiment, the host computer NPU has a built-in 3D Gaussian Splatting (3DGS) algorithm module. Unlike traditional mesh modeling, the system directly uses the 3DGS algorithm for training at the edge after acquiring eight multi-angle images.

[0046] In this embodiment, Gaussian splash point cloud data is generated, and the reconstruction time for a single sample is reduced to less than 8 seconds, which greatly improves the processing efficiency in the field.

[0047] In this embodiment, to ensure the system's stability in high-altitude areas such as Nagqu, Tibet, special designs have been implemented for the power supply and communication: In this embodiment, the power management system 3 integrates a 24V / 10Ah lithium battery pack, which is stepped down to 12V (powering the motor) and 5V (powering the motherboard / camera) by a DC-DC module.

[0048] In this embodiment, an intelligent power consumption strategy (power off idle modules and automatic screen sleep) is adopted, and the battery life is ≥12 hours when fully charged.

[0049] In this embodiment, the system integrates a 5G / WIFI communication module. In extreme environments where network signal strength is as low as -110dBm, the system will automatically cache the collected data (such as approximately 15GB of 200 sets of sample data) completely on the local solid-state drive.

[0050] In this embodiment, once the network is restored, the system automatically starts the incremental synchronization mechanism to continue transmitting data to the cloud server, ensuring zero data loss.

[0051] In this embodiment, the system has been field-tested in Nagqu, Tibet (at an altitude of over 4500 meters): The battery pack can maintain continuous operation for 10.5 hours at an ambient temperature of -5℃. It can complete standardized data collection (including 3D reconstruction) for over 2000 samples per day. The entire process is automated with a failure rate of less than 0.5%.

[0052] The above description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the concept described herein through the above teachings or related technologies or knowledge. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A fully automated Cordyceps sinensis photography system, comprising a closed collection chamber (1), a precision mechanical transmission module (5), a multimodal sensing module (7), an edge computing control unit (2), and a power management system (3); characterized in that, The enclosed acquisition chamber (1) is equipped with a shooting space. The precision mechanical transmission module (5) is located in the shooting space and is used to fix and rotate the Cordyceps sinensis sample. The multimodal sensing module (7) includes a high-definition camera (4) and a supplementary lighting unit (6). The edge computing control unit (2) includes a bottom-level MCU and a host computer NPU for controlling mechanical transmission and image acquisition. The multimodal sensing module (7) also includes a high-precision weighing sensor and a GPS positioning module. The edge computing control unit (2) is configured to synchronously acquire image, weight and location data in one acquisition cycle and encrypt and bind the data for storage.

2. The fully automatic photographing system for Cordyceps sinensis according to claim 1, characterized in that, The enclosed collection chamber (1) adopts an integrated handle design, with overall dimensions of 300mm long × 300mm wide × 300mm high, and a total weight of less than 6kg.

3. The fully automatic photographing system for Cordyceps sinensis according to claim 1, characterized in that, The precision mechanical transmission module (5) includes a stepper motor, a coupling and a sample fixing fixture; the sample fixing fixture is a detachable clamp, and the stepper motor drives the turntable to rotate 360° with a positioning accuracy of 0.1°.

4. The fully automatic photographing system for Cordyceps sinensis according to claim 1, characterized in that, The edge computing control unit (2) adopts a dual-core main control architecture of "MCU + NPU": The underlying MCU uses an STM32F4 series chip, which is responsible for real-time control of the stepper motor, supplementary light, and reading of weighing data. The host computer NPU uses an NVIDIA edge computing module (Jetson Nano) to handle image processing, human-computer interaction, 5G data encryption upload, and local storage management.

5. The fully automatic photographing system for Cordyceps sinensis according to claim 1, characterized in that, The high-precision weight sensor is integrated below the sample placement platform, with a weighing accuracy of no less than 0.01g; the system also includes a GPS positioning module for acquiring the latitude, longitude, and altitude information of the collection point.

6. The fully automatic photographing system for Cordyceps sinensis according to claim 1, characterized in that, The power management system (3) integrates a 24V lithium battery pack and, with the help of intelligent power consumption strategy, provides a battery life of ≥12 hours when fully charged; the data transmission supports 5G / WIFI communication and has the function of resuming interrupted transmission, and automatically caches data to the local solid-state drive when the network signal is weak.

7. The fully automatic photographing system for Cordyceps sinensis according to claim 1, characterized in that, The system has built-in automated control software. After the user scans the code to enter the sample ID, the system automatically controls the rotation, shooting (collecting 4-8 multi-angle images each time), weighing, and encrypting and packaging the image data with weight, GPS, and time metadata for synchronization.

8. A method for collecting Cordyceps sinensis data is based on a fully automatic Cordyceps sinensis photography system as described in any one of claims 1-7, characterized in that, Includes the following steps: S1 Enter the sample ID; S2 controls the activation of the supplementary lighting unit and drives the sample to rotate to a preset angle; S3 triggers the high-definition camera to acquire images; S4 reads data from the weighing sensor and GPS module, and binds and stores them with image data.

9. A method for collecting Cordyceps sinensis data according to claim 8, characterized in that, In step S4, if the network signal strength is lower than a preset threshold, the system automatically caches the data locally and performs incremental synchronization automatically after the signal is restored.

10. A method for collecting Cordyceps sinensis data according to claim 8, characterized in that, The acquired multi-angle images are reconstructed in three dimensions using the 3D Gaussian Splatting algorithm or the SfM algorithm to generate a digital three-dimensional model of Cordyceps sinensis.

Images