A Machine Vision-Based Dynamic Recognition Method and System for Edible Fungi

Abstract

This invention discloses a dynamic recognition method and system for edible fungi based on machine vision. The invention relates to the technical field of machine vision. It determines the camera's shooting path relative to the edible fungi based on their distribution location and the camera's current position. Based on the shooting path and the cultivation environment of the edible fungi, it determines the camera's dynamic recognition mode, improving the accuracy of edible fungi recognition. The method determines a set of detailed features of the edible fungi based on the recognition of multiple current images. Based on this set of detailed features, it determines the species and life cycle of the edible fungi. Based on the species, life cycle, and cultivation environment of the edible fungi, it determines the cultivation level, triggering optimization of the cultivation environment to ensure healthy growth of the edible fungi and maintain the accuracy of the cultivation level, thus optimizing the cultivation environment.

Description

Technical Field

[0001] This invention relates to the field of machine vision technology, and in particular to a dynamic identification method and system for edible fungi based on machine vision. Background Technology

[0002] With the development of technology, edible fungi have gradually become one of people's foods. They are cultivated in specific environments within corresponding edible fungi cultivation spaces. Edible fungi refer to large, edible mushrooms, commonly known as mushrooms. In existing technologies, edible fungi exist within edible fungi cultivation spaces and undergo different life cycles. Edible fungi have corresponding forms in different life cycles. Cameras capture edible fungi from a single angle, only controlling the form of the edible fungi in a single direction, ignoring the shooting path of the edible fungi and the consideration of the cultivation scene, resulting in low accuracy in the identification of edible fungi. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides a dynamic identification method and system for edible fungi based on machine vision.

[0004] This invention provides a machine vision-based dynamic identification method for edible fungi, comprising: The distribution of edible fungi relative to the edible fungi cultivation space was collected; The cultivation scenarios for edible fungi are determined based on the distribution location of the fungi and multiple environmental parameters of the cultivation space. These scenarios include indoor cultivation, outdoor cultivation, and forest cultivation. The camera's shooting path relative to the edible fungi is determined based on the distribution location of the edible fungi and the current location of the camera. The camera's dynamic recognition mode for the edible fungi is determined based on the shooting path of the edible fungi and the cultivation scene of the edible fungi. The dynamic recognition mode includes a standard recognition mode, an environmental adaptation recognition mode, and an intelligent recognition mode. In this dynamic recognition mode, the set of detailed features of edible fungi is determined based on the recognition of multiple current images of edible fungi, and the species and life cycle of edible fungi are determined based on the set of detailed features of edible fungi. The cultivation level of edible fungi is determined based on the type of edible fungi, their life cycle, and the cultivation environment of the cultivation space. Trigger the optimization of the cultivation scenario in the edible fungi cultivation space to ensure the good growth of edible fungi in the cultivation space.

[0005] This invention provides a machine vision-based dynamic identification system for edible fungi, which is applied to the aforementioned machine vision-based dynamic identification method for edible fungi.

[0006] Compared with the prior art, the beneficial effects of the present invention are: In this embodiment of the invention, the distribution location of edible fungi relative to the edible fungi cultivation space is collected using the method described in this embodiment; the cultivation scenario of edible fungi is determined based on the distribution location of edible fungi and multiple environmental parameters of the edible fungi cultivation space; the shooting path of the camera relative to the edible fungi is determined according to the distribution location of edible fungi and the current position of the camera; and the dynamic recognition mode of the camera for edible fungi is determined according to the shooting path of edible fungi and the cultivation scenario of edible fungi. This method takes into account both the shooting path of edible fungi and the cultivation scenario of edible fungi, thereby improving the accuracy of edible fungi recognition.

[0007] Therefore, in this dynamic recognition mode, the set of detailed features of edible fungi is determined based on the recognition of multiple current images of edible fungi. The species and life cycle of edible fungi are determined based on the set of detailed features. The cultivation level of edible fungi is determined based on the species, life cycle, and cultivation scenario of the cultivation space, triggering the optimization of the cultivation scenario of the cultivation space to ensure the good growth of edible fungi in the cultivation space. The introduction of the species and life cycle of edible fungi ensures the accuracy of the cultivation level of edible fungi and realizes the optimization of the cultivation scenario of the cultivation space. Attached Figure Description

[0008] Figure 1 This is a flowchart illustrating the dynamic identification method for edible fungi based on machine vision in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structural composition of the machine vision-based dynamic identification system for edible fungi in an embodiment of the present invention. Detailed Implementation

[0009] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0010] Please see Figure 1 and Figure 2 A machine vision-based dynamic identification method for edible fungi is applied to dynamic identification scenarios of edible fungi. This machine vision-based dynamic identification method for edible fungi includes: Step S11: Collect the distribution location of edible fungi relative to the edible fungi cultivation space; Step S12: Determine the cultivation scenario for edible fungi based on multiple environmental parameters of the distribution location of edible fungi and the cultivation space of edible fungi; Step S13: Determine the shooting path of the camera relative to the edible fungi based on the distribution location of the edible fungi and the current position of the camera; determine the dynamic recognition mode of the camera for the edible fungi based on the shooting path of the edible fungi and the cultivation scene of the edible fungi. Step S14: In this dynamic recognition mode, the set of detailed features of edible fungi is determined based on the recognition of multiple current images of edible fungi, and the type and life cycle of edible fungi are determined based on the set of detailed features of edible fungi. Step S15: Determine the cultivation level of edible fungi based on the type, life cycle, and cultivation environment of the edible fungi cultivation space; Step S16: Trigger the optimization of the cultivation scenario in the edible fungus cultivation space to ensure the good growth of edible fungi in the edible fungus cultivation space; In step S11, the distribution position of edible fungi relative to the edible fungi cultivation space is collected; In the specific implementation of this invention, the specific steps are as follows: S111: Collect a distribution map of the edible fungi cultivation space, and determine multiple cultivation areas based on the identification of the distribution map of the edible fungi cultivation space, so as to determine the spatial position of multiple cultivation areas relative to the edible fungi cultivation space. S112: Locate and detect multiple culture areas, and determine the culture area corresponding to the edible fungus based on the location detection of multiple culture areas, and take the culture area corresponding to the edible fungus as the target culture area; S113: Construct a corresponding coordinate system based on the edible fungus cultivation space, and mark the coordinate position of the target cultivation area. Determine the distribution position of the edible fungus relative to the edible fungus cultivation space based on the coordinate position of the target cultivation area and the distribution position of the edible fungus relative to the target cultivation area.

[0011] In the embodiments of this application, a distribution map of the edible fungus cultivation space is collected. At the same time, a high-definition camera or other image acquisition device is used to take photos or videos of the cultivation space from an appropriate angle to ensure that the images are clear, unobstructed, and cover the entire cultivation space.

[0012] Different culture regions are identified in the distribution map, defined based on physical structures (such as culture racks, incubators) or markers (such as color coding, labels); optionally, image processing techniques (such as edge detection, region segmentation) are used to identify different regions in the image; regions are classified according to their shape, size, color or other features to determine whether they belong to different culture regions; machine learning algorithms, such as cluster analysis, are involved to automatically identify and classify culture regions.

[0013] Assign a relative position to each culture region for subsequent tracking and analysis; at this point, select a reference point (such as a corner or center point of the culture space) as the origin of the coordinate system; define the X-axis and Y-axis (or Z-axis, if considering three-dimensional space) according to the layout of the culture space; measure or estimate the position of each culture region relative to the origin and record its coordinates; computer vision techniques, such as feature point matching or template matching, are used to accurately determine the position of the culture region.

[0014] Furthermore, within the identified multiple cultivation areas, the specific location and extent of each area are further detected to ensure accurate matching of edible fungi with the corresponding cultivation areas. Optionally, image processing techniques (such as edge detection and contour extraction) are used to precisely delineate the boundaries of each cultivation area. If the cultivation area has specific markings (such as QR codes, barcodes, or color encoding), corresponding decoding techniques or color recognition algorithms are used to locate these markings, thereby more accurately determining the location of the cultivation area. The target object (in this case, the cultivation area) is quickly and accurately detected in the image.

[0015] After determining the specific location and extent of each cultivation area, it is necessary to further identify the edible fungi within each cultivation area and match them with the corresponding cultivation area. Optionally, within each cultivation area, object detection or image segmentation techniques can be used to identify the presence and location of edible fungi. Based on the location information of the edible fungi, they can be matched with the nearest cultivation area, or their corresponding cultivation area can be determined according to preset rules (such as the number and location of the petri dish). Optionally, edible fungi recognition models can be involved. These edible fungi recognition models have been trained to identify different types of edible fungi and distinguish them based on their characteristics (such as shape, color, and texture).

[0016] Once the correspondence between edible fungi and cultivation areas is determined, these cultivation areas are marked as target cultivation areas for further detailed analysis and management. Optionally, each target cultivation area is assigned a unique identifier (such as ID number, name, etc.). Information about the target cultivation areas (such as location, types and quantities of edible fungi contained therein, etc.) is stored in a database for subsequent retrieval and analysis.

[0017] Specifically, suppose there is an edible mushroom cultivation room with multiple layers of cultivation racks inside. Each layer of the cultivation rack holds multiple petri dishes, and each petri dish contains a different type of edible mushroom. In order to determine the cultivation area corresponding to each edible mushroom and mark it as the target cultivation area, the operation is carried out according to step S112: use a high-definition camera to capture images of the cultivation room, and use edge detection and contour extraction technology in image processing software to identify the boundaries of each petri dish; assume that the petri dishes are all standard circles or squares and are arranged neatly, so they can be distinguished by shape and position information.

[0018] Within each petri dish, a target detection algorithm (such as YOLO) is used to identify the presence and location of edible fungi. Since each petri dish typically cultivates only one type of edible fungus, the identified fungus is directly matched with the petri dish it belongs to. For example, if a shiitake mushroom is identified in a petri dish, then that petri dish is marked as the shiitake mushroom cultivation area. Each identified cultivation area is assigned a unique ID number, and this information is stored in a database. For example, the first petri dish on the first layer of the cultivation rack is marked as ID001, and the type and quantity of shiitake mushrooms in it are also stored in the database. This allows for easy retrieval and referencing of this information in subsequent analysis and management.

[0019] Therefore, a coordinate system is constructed based on the edible fungus cultivation space, and the coordinate position of the target cultivation area is marked. The distribution position of the edible fungus relative to the edible fungus cultivation space is determined based on the coordinate position of the target cultivation area and the distribution position of the edible fungus relative to the target cultivation area. This approach takes into account both the coordinate position of the target cultivation area and the distribution position of the edible fungus relative to the target cultivation area, ensuring the accuracy of the distribution position of the edible fungus relative to the edible fungus cultivation space.

[0020] At this point, in order to accurately describe the location of each cultivation area and the edible fungi within the cultivation space, a unified coordinate system needs to be constructed. Optionally, a fixed reference point can be selected as the origin of the coordinate system. This point is usually a specific and immovable location within the cultivation space, such as a corner or center point of the cultivation room. The two main directions of the coordinate system (X-axis and Y-axis) are determined, usually by selecting the long and short sides of the cultivation space as the X-axis and Y-axis. If the cultivation space is three-dimensional (such as a multi-layer cultivation rack), the Z-axis also needs to be determined, usually by selecting a direction perpendicular to the ground.

[0021] In the constructed coordinate system, a unique coordinate position is assigned to each target culture region so that they can be accurately located later. At this time, measuring tools (such as tape measure, laser rangefinder) or image processing techniques (such as feature point matching) are used to determine the position of each target culture region relative to the origin. Based on the measurement results, a coordinate point is marked for each target culture region in the coordinate system.

[0022] After determining the coordinates of the target culture area, the distribution of edible fungi relative to the entire culture space is further determined. Optionally, within each target culture area, image processing techniques or manual measurement are used to determine the position of the edible fungi relative to that culture area (such as the offset relative to the center of the petri dish). Combining the coordinates of the target culture area and the offset of the edible fungi relative to that area, the position of the edible fungi in the overall coordinate system is calculated.

[0023] Specifically, suppose there is a two-dimensional edible fungus cultivation chamber with multiple layers of cultivation racks installed inside. Each layer of cultivation racks holds multiple petri dishes, and each petri dish contains different types of edible fungi. In order to determine the distribution of the edible fungi relative to the entire cultivation space.

[0024] Choose the lower left corner of the cultivation chamber as the origin of the coordinate system (0,0). The long side of the cultivation chamber is the X-axis (positive direction to the right), and the short side is the Y-axis (positive direction upward). Use a laser rangefinder to measure the position of each petri dish relative to the origin. For example, the coordinates of the first petri dish from left to right and from front to back on the first layer of the cultivation rack are (20,30), indicating that it is 20 units away from the origin in the X-axis direction and 30 units away in the Y-axis direction. In each petri dish, assume that the edible fungi are located at the center of the petri dish (or measure their offset according to the actual situation). Therefore, if the first petri dish contains shiitake mushrooms and the shiitake mushrooms are located at the center of the petri dish, then the position of the shiitake mushrooms in the overall coordinate system is (20,30). If the shiitake mushrooms are offset from the center of the petri dish, for example, offset 5 units to the right and offset 3 units upward, then the actual position of the shiitake mushrooms is (25,33). By following these steps, the distribution of each edible fungus relative to the entire cultivation space can be accurately determined, providing precise data support for subsequent analysis and management. This information is used to monitor the growth of edible fungi, optimize cultivation conditions, and predict harvest time.

[0025] In step S12, the cultivation scenario for edible fungi is determined based on the distribution location of the edible fungi and multiple environmental parameters of the edible fungi cultivation space; In the specific implementation of this invention, the specific steps are as follows: S121: In the edible fungus cultivation space, mark the environmental sampling locations of the edible fungus cultivation space, and determine the corresponding environmental parameters based on the detection of the environmental sampling locations of the edible fungus cultivation space, so as to collect multiple environmental parameters of the edible fungus cultivation space; S122: Collect the distribution location of edible fungi, determine the local culture area of ​​edible fungi based on the distribution location and colony area of ​​edible fungi, and determine the surrounding environmental parameters of edible fungi based on the environmental monitoring of the local culture area of ​​edible fungi. S123: Determine the first culture coefficient based on the surrounding environmental parameters of the edible fungus and the spatial area of ​​the edible fungus cultivation space; determine the second culture coefficient based on multiple environmental parameters and spatial area of ​​the edible fungus cultivation space; determine the cultivation scenario of the edible fungus based on the mapping relationship between the first culture coefficient, the second culture coefficient and the cultivation scenario, which is an indoor cultivation scenario, an outdoor cultivation scenario and a forest cultivation scenario.

[0026] In the embodiments of this application, key locations for collecting environmental parameters are rationally laid out within the edible fungus cultivation space to ensure a comprehensive and accurate reflection of the environmental conditions of the entire cultivation space. Optionally, multiple environmental sampling points are planned based on the shape and size of the cultivation space and the distribution of edible fungi. Clear markings (such as labels or signs) are used to mark each sampling point for subsequent environmental parameter collection. The selection of sampling points should consider key environmental factors required for edible fungus growth, such as temperature, humidity, light, and CO2 concentration. Sampling points should be evenly distributed to avoid over-concentration or omission of certain areas.

[0027] At each marked collection point, use professional environmental monitoring equipment to obtain accurate environmental parameter data; optionally, prepare environmental monitoring equipment such as temperature and humidity sensors, light intensity meters, CO2 detectors, etc.; conduct tests at each collection point according to predetermined time intervals or as needed; record the environmental parameter data of each collection point, including temperature, humidity, light intensity, CO2 concentration, etc.

[0028] Furthermore, the specific location of the edible fungi in the cultivation space is determined to provide basic data for subsequent analysis of the growth status and environmental adaptability of the edible fungi. At this time, image recognition technology or manual observation and recording methods are used to locate the edible fungi in the cultivation space. In image recognition technology, a high-resolution camera is used to take pictures of the cultivation space, and then the location of the edible fungi is automatically identified by image processing software. In manual observation and recording, the location of the edible fungi is directly marked on the petri dish or culture rack with a marker pen, or its coordinates relative to a certain fixed point are recorded.

[0029] Based on the distribution of edible fungi, small local culture zones are delineated for more precise analysis of their growth environment. Optionally, a suitable local culture zone can be defined centered on the edible fungus, taking into account factors such as colony size, shape, and distance between adjacent fungi. These local culture zones can be circular, square, or other shapes, and their size should be sufficient to encompass the edible fungus and its surrounding microenvironment. The division of local culture zones should accurately reflect the actual growth environment of the edible fungi; different species of edible fungi or those at different growth stages require different criteria for defining local culture zones.

[0030] Within the designated local cultivation area, environmental parameters are measured to determine the surrounding environment of the edible fungi. Optionally, professional environmental monitoring equipment, such as temperature and humidity sensors, light intensity meters, and gas detectors, can be used to conduct multi-point measurements within the local cultivation area. Each measurement point should be distributed as evenly as possible within the local cultivation area to ensure that the collected environmental parameters are representative. The environmental parameter data for each measurement point are recorded, and the average environmental parameter value of the local cultivation area is calculated.

[0031] Specifically, suppose there is a mushroom cultivation rack with multiple petri dishes on it, each inoculated with a different type of edible fungus. To determine the environmental parameters surrounding the mushrooms, a high-resolution camera was used to photograph the cultivation rack. Image processing software was used to automatically identify the location of the mushrooms in the photographs and mark them. The species, growth stage, and location coordinates of each mushroom were recorded.

[0032] Centered on each identified edible fungus, a circular local culture area with a diameter of 5 cm was delineated based on the size and shape of its colony. The edges of each local culture area were marked with a marker pen for subsequent environmental monitoring.

[0033] Using temperature and humidity sensors, light intensity meters, and gas detectors, multiple measurements were taken within each local cultivation area. Each measurement point was evenly distributed within the local cultivation area to ensure the representativeness of the collected environmental parameters. Environmental parameter data for each measurement point were recorded, and the average temperature, humidity, light intensity, and gas concentration values ​​for each local cultivation area were calculated. Through these operations, the surrounding environmental parameters of the edible fungi were successfully determined, providing an important basis for subsequent analysis of the growth status and environmental adaptability of the edible fungi. These parameter data are used to assess the growth performance of edible fungi under different environments and whether corresponding measures need to be taken to adjust and optimize the cultivation conditions.

[0034] Therefore, a first culture coefficient is determined based on the surrounding environmental parameters of the edible fungi and the spatial area of ​​the edible fungi cultivation space. A second culture coefficient is determined based on multiple environmental parameters and the spatial area of ​​the edible fungi cultivation space. The cultivation scenario of the edible fungi is determined based on the mapping relationship between the first culture coefficient, the second culture coefficient, and the cultivation scenario. This cultivation scenario includes indoor cultivation scenario, outdoor cultivation scenario, and forest cultivation scenario, which takes into account the overall consideration of the mapping relationship between the first culture coefficient, the second culture coefficient, and the cultivation scenario, and ensures the accuracy of the cultivation scenario of the edible fungi.

[0035] At this point, based on the surrounding environmental parameters of the edible fungi (such as temperature, humidity, light, gas concentration, etc.) and the spatial area of ​​the cultivation space, the adaptability of the edible fungi to the local cultivation area is comprehensively evaluated to determine the first cultivation coefficient. Optionally, data on the surrounding environmental parameters of the edible fungi, including temperature, humidity, light intensity, CO2 concentration, etc., are collected and analyzed. Based on the growth characteristics of the edible fungi and the ideal growth environment, thresholds or ranges for each environmental parameter are set. The matching degree between the actual environmental parameters and the ideal environmental parameters is calculated using methods such as weighted average method and fuzzy comprehensive evaluation method. Combining the spatial area of ​​the cultivation space, and considering factors such as space utilization and ventilation conditions, the matching degree is corrected to obtain the first cultivation coefficient.

[0036] Based on multiple environmental parameters of the edible fungi cultivation space (such as overall temperature and humidity, light distribution, gas exchange, etc.) and the space area, the environmental quality of the entire cultivation space is comprehensively evaluated to determine the second cultivation coefficient. Optionally, data on the overall environmental parameters of the cultivation space, including temperature and humidity distribution, light intensity distribution, CO2 concentration, etc., are collected and analyzed. The effects of factors such as ventilation conditions, light uniformity, and temperature gradient of the cultivation space on the growth of edible fungi are evaluated. Taking into account the space area, spatial layout, equipment configuration, and other factors, the overall environmental quality is comprehensively evaluated. Based on the comprehensive evaluation results, the second cultivation coefficient is determined.

[0037] Based on the mapping relationship between the first culture coefficient, the second culture coefficient, and the culture scenario, the most suitable culture scenario for the growth of edible fungi is determined; optionally, a culture scenario mapping relationship is established, that is, the culture scenarios corresponding to different combinations of culture coefficients; according to the specific values ​​of the first culture coefficient and the second culture coefficient, the corresponding culture scenario is found in the mapping relationship; considering factors such as the type of edible fungi, the growth stage, and market demand, the culture scenario is appropriately adjusted and optimized.

[0038] Specifically, suppose there is an edible mushroom cultivation room for cultivating shiitake and oyster mushrooms; in order to determine the most suitable cultivation environment for them, the first cultivation coefficient is determined: collect and analyze the surrounding environmental parameter data of shiitake and oyster mushrooms in the local cultivation area; set the ideal range of temperature, humidity, light intensity, and CO2 concentration; calculate the matching degree between the actual environmental parameters and the ideal range, and obtain the first cultivation coefficients of shiitake and oyster mushrooms as 0.85 and 0.90, respectively.

[0039] Determine the second culture coefficient: Collect and analyze the overall environmental parameter data of the culture room; evaluate factors such as ventilation conditions and light uniformity of the culture room; comprehensively evaluate the overall environmental quality in combination with the space area and equipment configuration of the culture room; determine the second culture coefficient of the culture room as 0.80.

[0040] A mapping relationship between cultivation scenarios was established. For example, when the first cultivation coefficient is ≥0.8 and the second cultivation coefficient is ≥0.75, an indoor cultivation scenario was selected. Based on the first and second cultivation coefficients of shiitake and oyster mushrooms, it was determined that both are suitable for growth in an indoor cultivation scenario. Considering the growth characteristics and market demand of shiitake and oyster mushrooms, the indoor cultivation scenario was appropriately adjusted and optimized, such as adjusting the temperature and humidity control strategy and optimizing the light distribution. Through the above operations, the most suitable cultivation scenario for shiitake and oyster mushrooms was successfully determined to be the indoor cultivation scenario, and corresponding cultivation strategies were formulated. These strategies help improve the growth quality and yield of edible fungi and meet market demand.

[0041] In one embodiment of this application, a culture scenario matching table is collected, which lists the culture scenarios corresponding to different ranges of the first culture coefficient and the second culture coefficient; the culture scenario matching table is shown in Table 1:

[0042] Suppose there is an edible fungus whose first culture coefficient, calculated from its surrounding environmental parameters and spatial area, is 0.85, and its second culture coefficient, calculated from its overall environmental parameters and spatial area, is 0.75. Comparing the calculated first and second culture coefficients with the matching table, it is found that they fall within the range of 0.7-0.89, so the corresponding culture scenario is an outdoor culture scenario.

[0043] In step S13, the shooting path of the camera relative to the edible fungi is determined based on the distribution location of the edible fungi and the current position of the camera, and the dynamic recognition mode of the camera for the edible fungi is determined based on the shooting path of the edible fungi and the cultivation scene of the edible fungi. In the specific implementation of this invention, the specific steps are as follows: S131: Mark the corresponding camera in the edible fungus cultivation space, and determine the current position of the camera based on the position detection of the camera and the edible fungus cultivation space, and determine the shooting area of ​​the camera relative to the edible fungus based on the distribution location of the edible fungus and the current position of the camera. S132: The camera's shooting path relative to the edible fungus is determined based on the path planning of the camera relative to the shooting area of ​​the edible fungus. The camera moves along the shooting path and shoots the edible fungus from different directions. S133: Based on the division of the shooting path of edible fungi, multiple shooting nodes are determined. At the same time, the cultivation scene of edible fungi is collected. Based on the combination of the cultivation scene of edible fungi and multiple shooting nodes, multiple shooting environment combinations are determined. Based on the recognition of multiple shooting environment combinations, corresponding recognition parameters are determined. Based on multiple recognition parameters and mode mapping relationship, the dynamic recognition mode of the camera for edible fungi is determined. The dynamic recognition mode includes standard recognition mode, environmental adaptation recognition mode and intelligent recognition mode.

[0044] In the embodiments of this application, corresponding cameras are marked in the edible fungus cultivation space, and the current position of the camera is determined based on the position detection of the camera and the edible fungus cultivation space. The shooting area of ​​the camera relative to the edible fungus is determined based on the distribution location of the edible fungus and the current position of the camera. This takes into account both the distribution location of the edible fungus and the current position of the camera, ensuring the accuracy of the shooting area of ​​the camera relative to the edible fungus.

[0045] At this point, each camera in the edible mushroom cultivation space should be clearly marked. This typically involves assigning a unique identifier (such as a number, name, etc.) to the camera and recording it in relevant documents or systems. The purpose of marking is to ensure that each camera can be accurately identified and controlled in subsequent operations and management. Optionally, multiple cameras are installed in the edible mushroom cultivation room to monitor the growth of edible mushrooms. Each camera is numbered, for example, Cam1, Cam2, Cam3, etc., and these numbers are clearly marked on the camera casing. A camera distribution map is created in a prominent location in the edible mushroom cultivation room or in the management system, marking the location and number of each camera so that managers can quickly find and identify each camera.

[0046] In this step, location detection technology is needed to determine the current location of the camera. This can be achieved in various ways, such as using a GPS positioning system (if the camera is outdoors and has a good signal), RFID technology (if the camera is equipped with an RFID tag and a corresponding reader), or visual SLAM (Simultaneous Localization and Mapping) technology (if the camera has visual perception capabilities). Optionally, in the edible mushroom cultivation room, since the cameras are installed in fixed indoor locations, RFID technology is used to determine the camera's location. Each camera is equipped with an RFID tag, and multiple RFID readers are installed in the edible mushroom cultivation room. When it is necessary to determine the camera's location, the information from the RFID tag is read, and the current location of the camera is determined by the location information from the reader.

[0047] Determining the camera's shooting area relative to the edible fungi is based on the distribution location of the fungi and the camera's current location. This typically involves calculating parameters such as the camera's field of view and focal length, and combining this with the distribution of the fungi to determine the shooting area. Optionally, the location of each camera and the distribution of the fungi are already known (e.g., through manual observation or previous monitoring systems). For each camera, the area it can capture is calculated based on its field of view (e.g., horizontal and vertical angles) and focal length. Then, this shooting area is compared with the distribution of the fungi to determine the actual shooting area of ​​the camera relative to the fungi.

[0048] Specifically, suppose there is a camera, Cam1, located in the upper left corner of a mushroom cultivation room. Using RFID technology, Cam1's current position is determined. Then, based on Cam1's field of view (e.g., 60 degrees horizontally and 45 degrees vertically) and focal length (e.g., 5 meters), the area that Cam1 can capture is calculated. Finally, this captured area is compared with the distribution of mushrooms in the cultivation room, revealing that Cam1 can capture a portion of the mushrooms in the upper left corner. Therefore, Cam1's captured area relative to the mushrooms is determined to be a portion of the upper left corner of the cultivation room. Through this detailed analysis and example, a better understanding of how to mark cameras in a mushroom cultivation space, determine the camera's current position, and define the camera's captured area relative to the mushrooms is crucial for subsequent accurate monitoring and management of the mushrooms.

[0049] Furthermore, the camera's shooting path relative to the edible fungus is determined based on the path planning of the camera relative to the shooting area of ​​the edible fungus. The camera moves along the shooting path and shoots the edible fungus from different directions, which takes into account the overall consideration of the path planning of the camera relative to the shooting area of ​​the edible fungus and ensures the accuracy of the camera's shooting path relative to the edible fungus.

[0050] At this point, path planning is performed based on the camera's shooting area relative to the edible fungus. The goal of path planning is to determine the optimal movement path of the camera within the shooting area to ensure that all parts of the edible fungus can be captured comprehensively and efficiently. This usually involves comprehensive consideration of factors such as the shape, size, obstacles, and distribution of the edible fungus in the shooting area. Path planning employs various algorithms, such as Dijkstra's algorithm, genetic algorithm, and particle swarm optimization algorithm. These algorithms generate optimal or suboptimal paths based on different objectives and constraints.

[0051] Optionally, information about the camera's shooting area relative to the edible fungi is first obtained through previous steps (such as S131); then, a map of the shooting area is drawn in two-dimensional or three-dimensional space based on the shape, size, and distribution of the edible fungi; next, a suitable path planning algorithm is selected, and corresponding objectives and constraints are set (such as shooting efficiency, shooting quality, camera movement speed, etc.); finally, the path planning algorithm is run to generate the optimal movement path of the camera within the shooting area.

[0052] Based on the path planning results, determine the specific shooting path of the camera. The shooting path should include the camera's starting point, ending point, and key points along the way (such as turning points, shooting points, etc.). These key points will guide how the camera moves and shoots during the shooting process. Optionally, extract the optimal movement path of the camera from the output of the path planning algorithm. Determine the camera's starting point and ending point based on the shape and length of the path. Set a series of key points on the path, which will serve as reference points for the camera during the shooting process. Based on shooting requirements (such as shooting angle, shooting distance, etc.), set and adjust each key point in detail.

[0053] The camera will move along a predetermined shooting path and capture images of edible fungi at key points. During the shooting process, the camera needs to adjust its shooting direction, focal length, and other parameters to ensure that clear and complete images of the edible fungi are captured. To achieve automatic camera movement and shooting, some automation technologies and equipment are required, such as robotics and automatic control systems. Optionally, the camera can be mounted on an automatically moving platform (such as a robot or unmanned vehicle). Then, based on the determined shooting path and key points, the camera's movement trajectory and shooting parameters are set. Next, the automatic control system is activated, causing the camera to move along the shooting path. During the movement, the camera will stop and capture images at each key point. While capturing images, the camera will automatically adjust its shooting direction, focal length, and other parameters based on its built-in sensors and control system. Finally, the captured images of the edible fungi are stored and processed for subsequent analysis and identification.

[0054] Specifically, suppose there is a camera, Cam2, in an edible mushroom cultivation room, responsible for filming the edible mushrooms in a certain area of ​​the room. A path has been planned based on Cam2's filming area, and its filming path has been determined. The filming path includes a starting point A, an ending point B, and several key points C, D, and E along the way. During the filming process, Cam2 first starts from the starting point A and moves towards the ending point B along the planned path. During its movement, Cam2 will stop at key point C and film, at which point it will adjust its filming direction to ensure that the edible mushrooms in that area are filmed. Then, Cam2 will continue to move and perform similar filming operations at key points D and E. Finally, Cam2 will reach the ending point B, completing the entire filming process. Through the detailed analysis and examples of the above steps, a better understanding of how to plan the path of the camera relative to the filming area of ​​the edible mushrooms and determine the camera's filming path is gained. This is crucial for achieving automatic camera movement and filming, as well as for subsequent accurate analysis and identification of the edible mushrooms.

[0055] Therefore, multiple shooting nodes are determined based on the division of the shooting path of edible fungi. At the same time, the cultivation scene of edible fungi is collected. Multiple shooting environment combinations are determined based on the combination of the cultivation scene of edible fungi and multiple shooting nodes. The corresponding recognition parameters are determined based on the recognition of multiple shooting environment combinations. Based on multiple recognition parameters and mode mapping relationship, the dynamic recognition mode of the camera for edible fungi is determined. This dynamic recognition mode includes standard recognition mode, environmental adaptation recognition mode and intelligent recognition mode, which is compatible with the overall consideration of the shooting path and cultivation scene of edible fungi, and improves the recognition accuracy of edible fungi.

[0056] At this point, multiple shooting nodes are defined based on the shooting path of the edible fungi. Shooting nodes are the locations where the camera stops and takes detailed pictures during the shooting process. The selection of these nodes should be able to fully cover the growth area of ​​the edible fungi, and take into account the growth characteristics of the edible fungi, shooting requirements, and environmental factors. In order to determine the shooting nodes, the shooting path is divided into several segments, and the end point or key point of each segment can be used as a shooting node. At the same time, the number of shooting nodes on the path is appropriately increased according to factors such as the distribution density and growth stage of the edible fungi.

[0057] Optionally, the camera's shooting path information is first obtained through previous steps (such as S132); then, the path is divided into several segments according to the shape and length of the shooting path and the distribution of edible fungi; next, shooting nodes are set at the end point or key point of each segment to ensure that these nodes can fully cover the growth area of ​​edible fungi; finally, the position and number of shooting nodes are fine-tuned according to shooting requirements and environmental factors.

[0058] Information on the cultivation environment of edible fungi is collected. This information includes environmental factors such as the growth environment, light conditions, temperature, and humidity of the edible fungi, as well as growth information such as the growth status and morphological characteristics of the fungi. This information will be used for subsequent analysis and identification. There are various methods for collecting information on the cultivation environment, such as using sensors for real-time monitoring and manual observation and recording. These methods are selected according to the actual situation.

[0059] Optionally, multiple sensors are installed in the edible fungus cultivation room to monitor environmental factors such as light, temperature, and humidity in real time. At the same time, dedicated personnel are arranged to conduct manual observation and recording of information such as the growth status and morphological characteristics of the edible fungi. The collected cultivation scene information will be stored in a database for subsequent analysis and identification.

[0060] Multiple shooting environment combinations are determined based on the cultivation scenario of edible fungi and the combination of multiple shooting nodes. The shooting environment combination refers to the different shooting environments formed under different shooting nodes due to different environmental factors and different growth states of edible fungi. In order to determine the shooting environment combination, each shooting node is combined with different environmental factors and growth states of edible fungi to form multiple shooting environments. Then, based on shooting requirements and recognition accuracy requirements, the optimal shooting environment combination is selected.

[0061] Optionally, based on the collected cultivation scene information, the environmental factors and edible fungi growth status under different shooting nodes are first determined; then, each shooting node is combined with different environmental factors and edible fungi growth status to form multiple shooting environment combinations; next, these shooting environment combinations are screened and optimized according to shooting requirements and recognition accuracy requirements; finally, the optimal shooting environment combination is determined for subsequent shooting and recognition work.

[0062] The identification parameters are determined based on the identification of multiple shooting environment combinations. The identification parameters refer to the feature parameters used to identify edible fungi, such as color, texture, and shape. These parameters will be used in subsequent image processing and recognition algorithms. In order to determine the identification parameters, the edible fungi images under each shooting environment combination are preprocessed and analyzed to extract the feature parameters used for identification. Then, based on factors such as the importance and stability of these feature parameters, the optimal identification parameters are selected.

[0063] Optionally, the edible fungus images under each shooting environment combination were preprocessed and analyzed, such as denoising and contrast enhancement; feature parameters such as color, texture, and shape were extracted from the images; these parameters were screened and optimized based on factors such as the importance and stability of the feature parameters; and the optimal recognition parameters were determined for use in subsequent image processing and recognition algorithms.

[0064] The dynamic recognition mode of the camera for edible fungi is determined based on multiple recognition parameters and pattern mapping relationships. The dynamic recognition mode refers to the camera's ability to automatically adjust its recognition strategy and method according to the recognition parameters and pattern mapping relationships under different shooting environment combinations during the shooting process. The dynamic recognition mode includes standard recognition mode, environmentally adaptive recognition mode, and intelligent recognition mode. The standard recognition mode refers to the recognition method used by the camera in a standard environment. The environmentally adaptive recognition mode refers to the camera's ability to automatically adjust the recognition parameters and methods according to changes in environmental factors. The intelligent recognition mode refers to the camera's ability to use advanced algorithms and machine learning technology for intelligent analysis and recognition.

[0065] Optionally, a dynamic recognition mode library for the camera is established based on the previously determined recognition parameters and mode mapping relationship; a suitable recognition mode is selected from the dynamic recognition mode library according to the changes in the shooting environment and recognition requirements; during the shooting process, the camera will automatically adjust its recognition strategy and method according to the selected recognition mode to achieve accurate recognition of edible fungi; finally, the recognition results of the camera are verified and optimized to improve the accuracy and stability of recognition.

[0066] Specifically, assuming there is a camera Cam3 in the edible mushroom cultivation room, it is responsible for filming the edible mushrooms in a specific area of ​​the cultivation room; multiple shooting nodes have been determined according to the shooting path of Cam3, and the cultivation scene information of edible mushrooms has been collected; then, based on the combination of shooting nodes and cultivation scene information, multiple shooting environment combinations are determined; under each shooting environment combination, the edible mushroom images are preprocessed and analyzed, and feature parameters such as color, texture, and shape are extracted, and the optimal recognition parameters are selected; next, a dynamic recognition mode library for Cam3 is established, including standard recognition mode, environmentally adapted recognition mode, and intelligent recognition mode.

[0067] During the shooting process, Cam3 selects a suitable recognition mode from its dynamic recognition mode library based on changes in the shooting environment and recognition requirements. For example, in environments with sufficient light and suitable temperature, Cam3 selects the standard recognition mode; in environments with insufficient light or excessively high / low temperatures, Cam3 selects the environmental adaptation recognition mode, automatically adjusting recognition parameters and methods according to changes in environmental factors; in complex or uncertain environments, Cam3 selects the intelligent recognition mode, utilizing advanced algorithms and machine learning techniques for intelligent analysis and recognition. Through detailed analysis and examples of the above steps, a better understanding of how to determine the camera's dynamic recognition mode based on the shooting path and cultivation scene of edible fungi is gained. This is of great significance for achieving intelligent shooting and recognition by the camera, as well as for the subsequent accurate analysis and management of edible fungi.

[0068] In step S14, in this dynamic recognition mode, a set of detailed features of edible fungi is determined based on the recognition of multiple current images of edible fungi, and the type and life cycle of edible fungi are determined based on the set of detailed features of edible fungi. In the specific implementation of this invention, the specific steps are as follows: S141: In this dynamic recognition mode, the camera performs multi-directional recognition of the edible fungus along the dynamic recognition mode to output multiple current images of the edible fungus and mark the environmental parameters in each current image; S142: Perform feature recognition on multiple current images, determine multiple detailed features based on the feature recognition of multiple current images, and determine multiple detailed features as morphological features of edible fungi in their corresponding life cycle; determine a set of detailed features of edible fungi based on multiple detailed features; S143: In the detailed feature set of edible fungi, a first detailed feature combination and a second detailed feature combination are determined based on the division of the detailed feature set of edible fungi. The species of edible fungi are determined based on the identification of the first detailed feature combination. The life cycle of edible fungi is determined based on the identification of the second detailed feature combination. The first detailed feature combination is a combination of local morphological features of edible fungi, and the second detailed feature combination is a combination of overall morphological features of edible fungi.

[0069] In the embodiments of this application, the camera first initializes itself according to preset or real-time calculated dynamic recognition mode parameters, including shooting angle, focal length, exposure time, etc., which will guide the camera on how to perform subsequent shooting work; at the same time, the camera then moves along a preset or dynamically planned shooting path, which is derived based on the growth characteristics of edible fungi, shooting requirements, and environmental factors; during the movement, the camera adjusts its shooting position and angle according to different nodes or areas on the path.

[0070] Upon reaching each shooting node or area, the camera performs multi-directional recognition. This means that the camera will not only capture the front or main part of the edible fungus, but also attempt to capture images from other angles (such as the side, top, etc.) to obtain more comprehensive image information. This multi-directional recognition helps to capture the morphological characteristics of the edible fungus from different perspectives, providing richer data for subsequent analysis and identification. After completing the shooting at each node, the camera will output multiple current images, including views from different angles such as the front, side, and top, as well as images under different lighting and focal length conditions. These images will serve as the basis for subsequent analysis and identification.

[0071] While outputting each image, the camera also marks or associates the current environmental parameters (such as light intensity, temperature, humidity, etc.) on the image. These environmental parameters are crucial for understanding the growth status and morphological characteristics of edible fungi, as well as for subsequent analysis and identification.

[0072] Specifically, suppose there is a smart camera in an edible mushroom cultivation room. It is equipped with a dynamic recognition mode function. The camera is set to move along a preset shooting path and perform multi-directional recognition at different nodes. The camera is first initialized according to preset parameters, including shooting angle, focal length, etc. The camera moves along the preset path in the edible mushroom cultivation room, which covers the main growth area of ​​edible mushrooms.

[0073] Upon reaching each shooting node, the camera captures front, side, and top views of the edible fungus. For example, at the first node, the camera captures a front view of the fungus, showing the complete shape and color of the cap. At the second node, the camera captures a side view of the fungus, showing the structure and distribution of the gills. At the third node, the camera captures a top view of the fungus, showing the morphology and size of the entire fungus. The camera saves the multi-directional recognition results (i.e., multiple current images) at each node as the basis for subsequent analysis and recognition. On each image, the camera also labels the environmental parameters at that time, such as light intensity, temperature, and humidity. For example, on the front view of the first node, information such as light intensity of 1000 lux, temperature of 25°C, and humidity of 80% is labeled.

[0074] Furthermore, multiple current images are preprocessed. Preprocessing includes image enhancement, denoising, and cropping to improve image quality and provide clear image data for subsequent feature recognition. Simultaneously, the system extracts features from the preprocessed images. Feature extraction is a crucial step in image analysis, aiming to extract useful information or features from images that describe the morphology, color, texture, and other attributes of edible fungi. In this step, the system utilizes computer vision algorithms, such as edge detection, corner detection, and texture analysis, to extract multiple detailed features of the edible fungi.

[0075] After extracting the features, the system determines the morphological characteristics of edible fungi in the corresponding life cycle based on these features. These morphological characteristics include the size, shape, and color of the cap, the density and structure of the gills, and the overall growth state of the fungus. The system then uses preset rules or machine learning models to determine whether these features belong to typical characteristics of a certain life cycle stage. Finally, the system integrates all extracted detailed features into a detailed feature set. This detailed feature set will contain multiple morphological characteristics of edible fungi at different life cycle stages, providing comprehensive data support for subsequent classification, identification, or monitoring.

[0076] Specifically, suppose we have a set of image data about a certain edible fungus at different stages of its life cycle; we enhance the images to improve their contrast and clarity; we remove noise and interfering information from the images, such as background clutter or shadows; and we crop the images to retain only the edible fungus portion in order to extract features more accurately.

[0077] Edge detection algorithms are used to extract the contour features of edible fungi, such as the shape and size of the cap; color space transformation and color histogram analysis are used to extract the color features of edible fungi, such as the color distribution and variation of the cap; and texture analysis algorithms are applied to extract the texture features of edible fungi, such as the density and structure of the gills.

[0078] Based on the extracted contour features, the life cycle stage of the edible fungus is determined; for example, a large and regularly shaped cap indicates that the fungus is in its mature stage. Color features are used to further confirm the species and growth status of the fungus; for example, specific color changes indicate that the fungus is about to enter its senescence stage. Texture features are used to help determine the health status and growth environment of the fungus; for example, dense and uniform gills indicate that the fungus is growing under good conditions. All extracted features are integrated into a detailed feature set, including the shape, size, and color of the cap, and the density and structure of the gills. This set will serve as the basis for subsequent analysis and identification, used for tasks such as monitoring the growth status of edible fungi, predicting their life cycle stage, or identifying species.

[0079] Therefore, based on the division of the detailed feature set of edible fungi, the first detailed feature combination and the second detailed feature combination are determined. The species of edible fungi are determined based on the identification of the first detailed feature combination, and the life cycle of edible fungi is determined based on the identification of the second detailed feature combination. The first detailed feature combination is a combination of local morphological features of edible fungi, and the second detailed feature combination is a combination of overall morphological features of edible fungi, thus introducing the species and life cycle of edible fungi.

[0080] At this point, the system will divide the previously constructed set of detailed features of edible fungi. This division is based on the nature and purpose of the features, with the aim of dividing the features into two groups: one group is used to identify the types of edible fungi (first detailed feature combination), and the other group is used to determine the life cycle of edible fungi (second detailed feature combination).

[0081] Introducing the first detailed feature set: The first detailed feature set mainly consists of the local morphological features of edible fungi. These features are usually closely related to the species of edible fungi, such as the specific shape of the cap, color pattern, and gill structure. The system will identify the species of edible fungi based on these local features.

[0082] Introducing a second set of detailed features: The second set of detailed features includes the overall morphological features of edible fungi. These features reflect the overall changes of edible fungi at different life cycle stages, such as the size of the fungus, color changes, and texture. The system will determine the life cycle stage of the edible fungi based on these overall features.

[0083] After determining the first and second detailed feature combinations, the system will use machine learning algorithms or classifiers to analyze and identify these features; for the first detailed feature combination, the system will output the identified edible fungus species; for the second detailed feature combination, the system will output the determined life cycle stage.

[0084] Specifically, suppose there is a set of detailed features about a certain unknown edible fungus. From the set of detailed features, select local features closely related to the species of edible fungus, such as the specific shape of the cap (e.g., umbrella-shaped, bell-shaped, etc.), color pattern (e.g., red, yellow, brown, etc.), and gill structure (e.g., dense, sparse, forked, etc.), to form the first set of detailed features. At the same time, select the overall features related to the life cycle of edible fungus, such as changes in the size of the fungus (from small to large), changes in color (from light to dark), and changes in texture (from fine to coarse), to form the second set of detailed features.

[0085] The system analyzes the first detailed feature combination and finds that the cap of the edible fungus is umbrella-shaped, the color pattern is yellow, and the gill structure is dense and branched. Based on these features, the system identifies the species of the edible fungus as "Golden Umbrella Mushroom".

[0086] The system analyzes the second set of detailed features and finds that the edible fungus is of moderate size, its color gradually changes from light yellow to dark yellow, and its texture changes from fine to slightly coarse. Based on these changes in overall characteristics, the system determines that the edible fungus is currently in its mature stage. Ultimately, the system identifies the edible fungus as "Golden Umbrella Mushroom" and determines that it is in its mature stage. Through the above steps and examples, the system clearly explains the process of dividing the detailed feature set, determining the first and second sets of detailed features, and identifying the species and life cycle in step S143. This helps to better utilize detailed feature sets to analyze and identify the species and life cycle of edible fungi.

[0087] In one embodiment of this application, a lifecycle matching table is collected, as shown in Table 2:

[0088] Now, a set of features has been extracted: the cap is round with wavy edges, and the gills are densely branched; the body is medium-sized, golden in color, but the texture is slightly rough (due to environmental factors); these features are compared with the entries in the matching table, and it is found that they best match the maturity characteristics of "Golden Umbrella Mushroom"; therefore, the edible fungus is identified as "Golden Umbrella Mushroom" and is in the maturity stage; however, it should be noted that the roughness of the texture is slightly different from that in the table.

[0089] In step S15, the cultivation level of edible fungi is determined based on the type of edible fungi, their life cycle, and the cultivation scenario of the cultivation space. In the specific implementation of this invention, the specific steps are as follows: S151: Collect the types and life cycle of edible fungi, determine the first cultivation level coefficient based on the types of edible fungi and the cultivation scenario of the edible fungi cultivation space, and determine the second cultivation level coefficient based on the life cycle of edible fungi and the cultivation scenario of the edible fungi cultivation space. S152: Determine the culture level of edible fungi based on the first culture level coefficient, the second culture level coefficient, and the culture level mapping relationship.

[0090] In the embodiments of this application, the types and life cycles of edible fungi are collected, and a first cultivation level coefficient is determined based on the types of edible fungi and the cultivation scenario of the edible fungi cultivation space. A second cultivation level coefficient is determined based on the life cycle of edible fungi and the cultivation scenario of the edible fungi cultivation space. This approach takes into account the overall consideration of the types of edible fungi and the cultivation scenario of the edible fungi cultivation space, ensuring the accuracy of the first cultivation level coefficient.

[0091] At this point, accurately recording the species information of the edible fungi (such as scientific name, common name, etc.) and their life cycle stage (such as spore stage, mycelial stage, fruiting body stage, etc.) is the basis for determining the cultivation level coefficient. Determining the first cultivation level coefficient: Based on the collected edible fungi species and the cultivation environment (including environmental factors such as temperature, humidity, light, and ventilation), it is necessary to use preset rules or models to calculate the first cultivation level coefficient. This coefficient reflects the growth potential and adaptability of the edible fungi species under the current cultivation environment. Usually, a cultivation environment more suitable for the growth of the species will yield a higher first cultivation level coefficient.

[0092] Determining the second cultivation level coefficient: In addition to considering the type of edible fungus and the cultivation environment, it is also necessary to determine the second cultivation level coefficient in combination with the life cycle of the edible fungus. Edible fungi at different life cycle stages have different requirements for environmental factors. Therefore, it is necessary to adjust the cultivation environment according to these requirements and calculate the second cultivation level coefficient accordingly. This coefficient also reflects the growth status and adaptability of the type of edible fungus at the current life cycle stage under the current cultivation environment.

[0093] Specifically, suppose there is an edible fungus called "oyster mushroom" that is currently in the early stage of fruiting body development; the cultivation environment is set to a temperature of 20°C, humidity of 85%, moderate light, and good ventilation; the species of edible fungus is recorded as "oyster mushroom" and the life cycle stage is "early stage of fruiting body development".

[0094] Determine the first cultivation level coefficient: It is known that "oyster mushrooms" grow best in an environment with a temperature of 20-25°C and a humidity of 80%-90%; therefore, the current cultivation scenario (temperature 20°C, humidity 85%) is very suitable for the growth of "oyster mushrooms"; based on this information, a high first cultivation level coefficient is calculated, such as 0.9 (assuming the coefficient range is 0-1, with 1 indicating the most suitable).

[0095] Determining the second cultivation level coefficient: Furthermore, it is known that *Pleurotus ostreatus* has moderate requirements for light and ventilation in the early stages of fruiting body development; the current cultivation scenario (moderate light, good ventilation) also meets these requirements; therefore, a similarly high second cultivation level coefficient is calculated, such as 0.85 (again assuming the coefficient range is 0-1); in summary, the first cultivation level coefficient for *Pleurotus ostreatus* under the current cultivation scenario is 0.9, and the second cultivation level coefficient is 0.85. These coefficients will be used in subsequent steps to determine the cultivation level of the edible fungus and serve as a basis for evaluating whether the cultivation scenario needs optimization. It should be noted that the rules or models for determining cultivation level coefficients in actual applications are more complex, involving the weighted summation of multiple environmental factors, the application of machine learning algorithms, etc. In addition, different types of edible fungi and different life cycle stages require different rules and models to calculate the cultivation level coefficients; therefore, flexible adjustments are needed based on the actual situation during implementation.

[0096] Furthermore, the cultivation level of edible fungi is determined based on the first cultivation level coefficient, the second cultivation level coefficient, and the cultivation level mapping relationship. This approach takes into account the overall considerations of the first cultivation level coefficient, the second cultivation level coefficient, and the cultivation level mapping relationship, ensuring the accuracy of the cultivation level of edible fungi.

[0097] At this point, the first and second cultivation level coefficients mentioned above are collected and combined with the preset cultivation level mapping relationship to determine the cultivation level of edible fungi. The cultivation level mapping relationship is usually a preset table or model that maps the two level coefficients to a specific cultivation level. This level is a simple classification (such as low, medium, and high) and also a specific value or range.

[0098] Specifically, suppose there is an edible fungus called "shiitake mushroom" that is currently in the middle stage of fruiting body development; in step S151, the first cultivation level coefficient of "shiitake mushroom" under the current cultivation scenario has been calculated to be 0.8 and the second cultivation level coefficient is 0.75. Now, there is a preset cultivation level mapping relationship. Based on this cultivation level mapping relationship, the first cultivation level coefficient of "shiitake mushroom" (0.8) and the second cultivation level coefficient of "shiitake mushroom" (0.75) are respectively mapped to the range of "medium" level. Therefore, the cultivation level of "shiitake mushroom" in the current cultivation scenario is determined to be "medium".

[0099] In step S16, the optimization of the cultivation scenario in the edible fungus cultivation space is triggered to ensure the good growth of edible fungi in the edible fungus cultivation space; A preset cultivation level threshold is determined based on the matching of the type and life cycle of edible fungi. If the cultivation level of edible fungi is lower than the preset cultivation level threshold, the cultivation scenario of the edible fungi cultivation space is optimized. At this time, the theoretical cultivation scenario is determined according to the type and life cycle of edible fungi, and the cultivation scenario of the edible fungi cultivation space is gradually optimized to the theoretical cultivation scenario in order to achieve good growth of edible fungi in the edible fungi cultivation space. The preset cultivation level threshold is determined based on the matching of the species and life cycle of the edible fungi. This threshold is one or more specific cultivation level values ​​used to assess whether the growth status of the edible fungi under the current cultivation conditions meets the standards. The determination of the threshold is usually based on historical data, expert experience or industry standards.

[0100] At this point, the actual cultivation level of the edible fungi is compared with the preset cultivation level threshold. If the actual cultivation level is lower than the preset threshold, it means that the current cultivation scenario is not conducive to the growth of the edible fungi and the optimization process needs to be triggered. Once it is determined that the cultivation scenario needs to be optimized, the theoretical cultivation scenario needs to be determined based on the type and life cycle of the edible fungi. The theoretical cultivation scenario refers to the combination of environmental conditions most suitable for the growth of this type of edible fungi at the current life cycle stage. These conditions include multiple aspects such as temperature, humidity, light, ventilation, and pH value.

[0101] Once the theoretical cultivation scenario is determined, the current cultivation scenario needs to be gradually adjusted to this theoretical value. This process needs to be carried out in stages to avoid sudden environmental changes causing excessive stress to the edible fungi. After each adjustment, the growth response of the edible fungi needs to be observed, and fine-tuning should be made according to the actual situation until the theoretical cultivation scenario is reached or approached.

[0102] Specifically, a preset cultivation level threshold for shiitake mushrooms in the mid-stage of fruiting body development is determined. Based on historical data and expert experience, it is assumed that when the cultivation level of shiitake mushrooms reaches "medium" or higher in the mid-stage of fruiting body development, both yield and quality are acceptable. Therefore, the preset cultivation level threshold is set to "medium." In summary, the cultivation level of shiitake mushrooms in the current cultivation scenario is determined to be "medium," and this level meets the requirement of the preset cultivation level threshold "medium." This means that the current cultivation conditions are suitable for the growth of shiitake mushrooms in the mid-stage of fruiting body development, and no additional cultivation scenario optimization is required. However, if the cultivation level is lower than the preset threshold (e.g., "low"), the cultivation scenario needs to be adjusted to improve the growth of the edible fungi.

[0103] Suppose there is an edible fungus called "enoki mushroom" that is currently in the late stage of fruiting body development. In step S152, it has been determined that the cultivation level of "enoki mushroom" in the current cultivation scenario is "low", while the preset cultivation level threshold is "medium". Therefore, it is necessary to trigger the optimization process of the cultivation scenario.

[0104] Comparing the actual cultivation level of "enoki mushrooms" (low) with the preset threshold of "medium", it was found that the actual cultivation level was lower than the preset threshold, thus triggering the optimization process. Based on the species of "enoki mushrooms" and the needs of the later stages of fruiting body development, the theoretical cultivation scenario was determined to be: temperature 15-18°C, humidity 85%-90%, weak diffused light, good ventilation, and a suitable pH range.

[0105] The current cultivation environment is gradually adjusted. First, the temperature is lowered from 20°C to 17°C, and the growth response of the enoki mushrooms is observed. If the growth improves, the humidity, light, and ventilation conditions are fine-tuned until they approach the theoretical cultivation environment. If the growth does not improve or even worsens, the accuracy of the theoretical cultivation environment needs to be reassessed, and corresponding adjustments made. Assuming that after a series of fine-tuning steps, the cultivation environment is finally adjusted to: 16°C, 88% humidity, weak diffused light, good ventilation, and a suitable pH value, the growth of the enoki mushrooms is significantly improved, and the cultivation level is upgraded to "medium" or higher. In summary, through the optimization process of step S153, the cultivation environment is gradually adjusted to the most suitable conditions according to the actual growth status and needs of edible fungi, thereby achieving good growth of edible fungi in the cultivation space. This process requires meticulous observation, accurate judgment, and timely adjustments to ensure the maximization of optimization effects.

[0106] In one embodiment of this application, a theoretical training scenario matching table is collected, as shown in Table 3:

[0107] Assuming the edible fungus is "enoki mushroom" and is in the late stage of fruiting body development, the corresponding theoretical cultivation scenario is found in the theoretical cultivation scenario matching table. The actual cultivation scenario is gradually adjusted to the theoretical cultivation scenario, which is achieved by adjusting environmental parameters such as temperature, humidity, light and ventilation. After each adjustment, the growth response of the edible fungus needs to be observed and fine-tuned according to the actual situation until the theoretical cultivation scenario is reached or close to it.

[0108] Please see Figure 2 , Figure 2 This is a schematic diagram of the structural composition of a machine vision-based dynamic identification system for edible fungi according to an embodiment of the present invention; the machine vision-based dynamic identification system for edible fungi includes: The acquisition module 21 is used to acquire the distribution position of edible fungi relative to the edible fungi cultivation space; The cultivation scenario module 22 is used to determine the cultivation scenario of edible fungi based on the distribution location of edible fungi and multiple environmental parameters of the edible fungi cultivation space; The dynamic recognition mode module 23 is used to determine the shooting path of the camera relative to the edible fungi based on the distribution location of the edible fungi and the current position of the camera, and to determine the dynamic recognition mode of the camera for the edible fungi based on the shooting path of the edible fungi and the cultivation scene of the edible fungi. The edible fungus identification module 24 is used to determine the set of detailed features of edible fungi based on the identification of multiple current images of edible fungi in the dynamic identification mode, and to determine the type and life cycle of edible fungi based on the set of detailed features of edible fungi. The cultivation level module 25 is used to determine the cultivation level of edible fungi based on the type, life cycle, and cultivation scenario of the edible fungi cultivation space. Optimization module 26 is used to trigger the optimization of the cultivation scenario in the edible fungi cultivation space to ensure the good growth of edible fungi in the edible fungi cultivation space.

[0109] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A dynamic identification method for edible fungi based on machine vision, characterized in that, include: The distribution of edible fungi relative to the edible fungi cultivation space was collected; The cultivation scenario for edible fungi is determined based on the distribution location of the fungi and multiple environmental parameters of the cultivation space. This cultivation scenario includes indoor, outdoor, and forest cultivation scenarios. The process involves: marking environmental sampling locations within the cultivation space and determining corresponding environmental parameters based on the detection of these locations; collecting multiple environmental parameters of the cultivation space; collecting the distribution location of the edible fungi; determining local cultivation areas based on the distribution location and colony area; determining surrounding environmental parameters based on environmental detection of the local cultivation area; determining a first cultivation coefficient based on the surrounding environmental parameters and the spatial area of ​​the cultivation space; determining a second cultivation coefficient based on the multiple environmental parameters and the spatial area of ​​the cultivation space; and finally, determining the cultivation scenario based on the mapping relationship between the first and second cultivation coefficients and the cultivation scenario. The camera's shooting path relative to the edible fungi is determined based on the distribution location of the edible fungi and the current location of the camera. The camera's dynamic recognition mode for the edible fungi is determined based on the shooting path of the edible fungi and the cultivation scene of the edible fungi. The dynamic recognition mode includes a standard recognition mode, an environmental adaptation recognition mode, and an intelligent recognition mode. In this dynamic recognition mode, the set of detailed features of edible fungi is determined based on the recognition of multiple current images of edible fungi, and the species and life cycle of edible fungi are determined based on the set of detailed features of edible fungi. The cultivation level of edible fungi is determined based on the type of edible fungi, their life cycle, and the cultivation environment of the cultivation space. Trigger the optimization of the cultivation scenario in the edible fungi cultivation space to ensure the good growth of edible fungi in the cultivation space.

2. The dynamic identification method for edible fungi based on machine vision according to claim 1, characterized in that, The distribution location of the collected edible fungi relative to the edible fungi cultivation space includes: A distribution map of edible fungi cultivation space was collected, and multiple cultivation areas were identified based on the distribution map of edible fungi cultivation space to determine the spatial location of multiple cultivation areas relative to the edible fungi cultivation space. Multiple cultivation areas are located and detected, and the cultivation area corresponding to the edible fungus is determined based on the location detection of multiple cultivation areas, and the cultivation area corresponding to the edible fungus is taken as the target cultivation area; A coordinate system is constructed based on the edible fungus cultivation space, and the coordinate position of the target cultivation area is marked. The distribution position of the edible fungus relative to the edible fungus cultivation space is determined based on the coordinate position of the target cultivation area and the distribution position of the edible fungus relative to the target cultivation area.

3. The dynamic identification method for edible fungi based on machine vision according to claim 1, characterized in that, The process of determining the camera's shooting path relative to the edible fungi based on their distribution location and the camera's current location, and determining the camera's dynamic recognition mode for the edible fungi based on their shooting path and cultivation environment, includes: The corresponding cameras are marked in the edible fungus cultivation space, and the current position of the cameras is determined based on the position detection of the cameras and the edible fungus cultivation space. The shooting area of ​​the cameras relative to the edible fungus is determined based on the distribution of the edible fungus and the current position of the cameras. The camera's shooting path relative to the edible fungus is determined based on the path planning of the camera relative to the shooting area of ​​the edible fungus. The camera moves along the shooting path and shoots the edible fungus from different directions.

4. The dynamic identification method for edible fungi based on machine vision according to claim 3, characterized in that, The camera's shooting path relative to the edible fungi is determined based on the distribution location of the edible fungi and the current position of the camera. The camera's dynamic recognition mode for the edible fungi is determined based on the shooting path and the cultivation environment of the edible fungi. This also includes: Multiple shooting nodes are determined based on the division of the shooting path for edible fungi. At the same time, the cultivation scene of edible fungi is collected. Multiple shooting environment combinations are determined based on the combination of the cultivation scene of edible fungi and multiple shooting nodes. The corresponding recognition parameters are determined based on the recognition of multiple shooting environment combinations. The dynamic recognition mode of the camera for edible fungi is determined based on multiple recognition parameters and the pattern mapping relationship.

5. The dynamic identification method for edible fungi based on machine vision according to claim 1, characterized in that, In this dynamic recognition mode, a set of detailed features of edible fungi is determined based on the recognition of multiple current images of the fungi. The species and life cycle of the edible fungi are then determined based on this set of detailed features, including: In this dynamic recognition mode, the camera performs multi-directional recognition of the edible fungus along the dynamic recognition mode to output multiple current images of the edible fungus and mark the environmental parameters in each current image; Feature recognition is performed on multiple current images, and multiple detailed features are determined based on the feature recognition of multiple current images. These multiple detailed features are the morphological features of edible fungi in their corresponding life cycle. A set of detailed features of edible fungi is determined based on these multiple detailed features.

6. The dynamic identification method for edible fungi based on machine vision according to claim 5, characterized in that, In this dynamic recognition mode, the detailed feature set of edible fungi is determined based on the recognition of multiple current images of the fungi, and the species and life cycle of the edible fungi are determined based on the detailed feature set. The method also includes: Based on the detailed feature set of edible fungi, a first detailed feature combination and a second detailed feature combination are determined. The species of edible fungi are determined based on the identification of the first detailed feature combination, and the life cycle of edible fungi is determined based on the identification of the second detailed feature combination. The first detailed feature combination is a combination of local morphological features of edible fungi, and the second detailed feature combination is a combination of overall morphological features of edible fungi.

7. The dynamic identification method for edible fungi based on machine vision according to claim 1, characterized in that, The determination of the cultivation level of edible fungi based on the type, life cycle, and cultivation environment of the edible fungi cultivation space includes: Collect information on the types and life cycles of edible fungi, determine the first cultivation level coefficient based on the types of edible fungi and the cultivation scenario of the edible fungi cultivation space, and determine the second cultivation level coefficient based on the life cycle of edible fungi and the cultivation scenario of the edible fungi cultivation space. The cultivation level of edible fungi is determined based on the first cultivation level coefficient, the second cultivation level coefficient, and the cultivation level mapping relationship.

8. The method for dynamic identification of edible fungi based on machine vision according to claim 1, characterized in that, The optimization of the cultivation scenario in the edible fungus cultivation space, to ensure the good growth of edible fungi in the cultivation space, includes: A preset cultivation level threshold is determined based on the matching of the edible fungi species and their life cycle. If the cultivation level of the edible fungi is lower than the preset cultivation level threshold, the cultivation scenario of the edible fungi cultivation space is optimized. At this time, the theoretical cultivation scenario is determined according to the edible fungi species and their life cycle, and the cultivation scenario of the edible fungi cultivation space is gradually optimized to the theoretical cultivation scenario in order to achieve good growth of edible fungi in the edible fungi cultivation space.

9. A dynamic identification system for edible fungi based on machine vision, characterized in that, The machine vision-based dynamic identification system for edible fungi is applied to the machine vision-based dynamic identification method for edible fungi as described in any one of claims 1-8.

Images