Construction method and device of animal original image library, equipment, medium and product

Abstract

The invention discloses an animal original image library construction method and device, equipment, a medium and a product, and relates to the field of pictures, and the method comprises the steps: obtaining a specimen set corresponding to each species in a target class group; the specimen set corresponding to the species comprises specimens of the species collected in various places in various habitats; the target class group is a medium-sized soil animal, a small-sized soil animal, a large-sized soil animal or an arthropod; for any one specimen, respectively obtaining pictures of the back surface, the ventral surface and the side surface of the specimen at each rotation angle, and obtaining a picture library corresponding to the specimen; and constructing the image libraries corresponding to all the specimens in the specimen set corresponding to each species in the target class group into the animal original image library. The animal original image library construction method and device can solve the defect that the image quality of an existing animal original database is poor.

Description

Technical Field

[0001] This application relates to the field of images, and in particular to a method, apparatus, device, medium, and product for constructing an original image library of animals. Background Technology

[0002] Currently, there are two methods for species identification. The first is morphological identification, which relies on taxonomists to observe microscopic morphological features such as the reproductive plate and bristles under a microscope. This method suffers from low efficiency, long processing time, and poor reproducibility. The second method is DNA barcoding technology. Although this method overcomes the limitations of phenotypic plasticity, it suffers from a high failure rate for identification in under-researched groups (such as rare species) due to low coverage of public databases.

[0003] Artificial intelligence-driven image recognition technology has provided a new paradigm for species identification, achieving an accuracy rate of over 95% in groups such as butterflies and birds. However, research on intelligent identification of small, medium, and large soil animals, as well as other smaller arthropods, lags far behind, with the core limiting factor being the poor image quality in existing animal image databases.

[0004] Therefore, a method for constructing an animal original image database is needed to address the shortcomings of poor image quality in existing animal original databases. Summary of the Invention

[0005] The purpose of this application is to provide a method, apparatus, device, medium, and product for constructing an animal original image database, which can solve the defect of poor image quality in existing animal original databases.

[0006] To achieve the above objectives, this application provides the following solution:

[0007] Firstly, this application provides a method for constructing an original animal image library, comprising:

[0008] Obtain the specimen collection corresponding to each species in the target group; the specimen collection corresponding to each species includes specimens of species collected from various locations in various habitats; the target group is a medium-sized soil animal, a small soil animal, a large soil animal, or an arthropod.

[0009] For any specimen, images of the back, ventral, and lateral sides of the specimen are acquired at various rotation angles to obtain the corresponding image library for the specimen;

[0010] The original animal image database is constructed by creating an image library of all specimens in the specimen collections corresponding to each species in the target group.

[0011] Secondly, this application provides an apparatus for constructing an animal original image library, comprising:

[0012] The specimen acquisition module is used to acquire the specimen set corresponding to each species in the target group; the specimen set corresponding to each species includes specimens of the species collected from various locations in various habitats; the target group is medium-sized soil animals, small soil animals, large soil animals, or arthropods;

[0013] The image acquisition module is used to acquire images of the back, ventral and lateral sides of any specimen at various rotation angles, and obtain the image library corresponding to the specimen.

[0014] The image library construction module is used to construct an original animal image library from the image libraries corresponding to all specimens in the specimen collections of each species in the target group.

[0015] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for constructing an animal original image library.

[0016] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method for constructing an animal original image library.

[0017] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method for constructing an animal original image library.

[0018] According to the specific embodiments provided in this application, this application has the following technical effects:

[0019] This application provides a method, apparatus, device, medium, and product for constructing an animal original image library. By acquiring species specimens from multiple habitats and sampling sites, this application addresses the problem of poor image quality caused by low coverage of morphological variations within the same species, which is predominantly derived from laboratory specimens in existing animal original image libraries. Furthermore, by acquiring images from multiple rotational angles, it overcomes the problem of poor image quality caused by the single image angle in existing animal original image libraries. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart illustrating a method for constructing an animal original image library according to an embodiment of this application;

[0022] Figure 2 A schematic diagram of the functional modules of an animal original image library construction device provided in an embodiment of this application;

[0023] Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0026] This application provides a method for constructing an original animal image library, such as... Figure 1 The following are included:

[0027] Step 201: Obtain the specimen set corresponding to each species in the target group; the specimen set corresponding to each species includes specimens of species collected from various locations in various habitats; the target group is medium-sized soil animals, small soil animals, large soil animals, or arthropods. Through systematic sampling across regions, multiple habitats, and diverse locations, the morphological representation bias caused by single sources is overcome.

[0028] Step 202: For any specimen, acquire images of the back, ventral, and lateral sides of the specimen at various rotation angles to obtain the corresponding image library.

[0029] Step 203: Construct an original animal image database by creating an image library containing all specimens from the specimen collections corresponding to each species in the target group. This original animal image database can be used for intelligent identification of the target group.

[0030] In another exemplary embodiment of this application, for any specimen, images of the dorsal, ventral, and lateral surfaces of the specimen are acquired at various rotation angles to obtain a corresponding image library for the specimen. This process further includes:

[0031] Precise morphological identification led by taxonomists. Accurate species identification is the foundation and prerequisite for highly precise intelligent identification of medium-sized, small, large, or arthropod soil animals. This requires taxonomists to employ morphological identification methods to accurately classify specimens based on morphological characteristics. Different medium-sized, small, large, or arthropod soil animals have different classification methods and systems. There is no unified standard, method, or procedure; therefore, taxonomists must classify and identify them according to the specific morphological classification methods for each group to ensure the authority and reliability of species annotation.

[0032] In another exemplary embodiment of this application, for any specimen, images of the dorsal, ventral, and lateral surfaces of the specimen are acquired at various rotation angles to obtain an image library corresponding to the specimen, specifically including:

[0033] Using a stereomicroscope equipped with a ring-shaped LED cold light source, with the entire specimen positioned within the computer-controlled imaging system, shadowless imaging was performed on the dorsal, ventral, and lateral surfaces of the specimen at various rotation angles, resulting in images of the dorsal, ventral, and lateral surfaces of the specimen at each rotation angle.

[0034] Images of the dorsal, ventral, and lateral surfaces of the specimen at various rotation angles are named according to the specimen parameters and image number to obtain the corresponding image library for the specimen. The specimen parameters include the specimen acquisition time, specimen acquisition location, specimen acquisition habitat, specimen acquisition plot, and the species name corresponding to the specimen.

[0035] In another exemplary embodiment of this application, the image library corresponding to all specimens in the specimen collections corresponding to each species in the target group is constructed into an original animal image library, specifically including:

[0036] Store images of specimens with identical parameters in the same folder.

[0037] For any folder, calculate the structural similarity between any two images in that folder.

[0038] If the structural similarity between two images is greater than a set threshold, then either of the two images is deleted to obtain the folder after removing redundancy; the set threshold is greater than 95%.

[0039] All folders, after removing redundancy, will be used to build a library of original animal images.

[0040] In another exemplary embodiment of this application, the rotation angles are 30 degrees, 60 degrees, and 90 degrees.

[0041] In another exemplary embodiment of this application, the image resolution of the stereomicroscope is greater than or equal to 20 million pixels, 40≤r≤110, and r is the magnification of the stereomicroscope.

[0042] In another exemplary embodiment of this application, obtaining the specimen set corresponding to each species in the target group specifically includes:

[0043] For any given species, animal samples corresponding to that species are collected from various locations in different habitats.

[0044] The collected animal samples were placed in specimen bottles containing 75% alcohol to obtain a collection of specimens corresponding to the species.

[0045] Currently, intelligent recognition of small, medium, and large soil animals, as well as other small arthropods, is limited by problems such as lack of image data, incomplete morphological feature collection, and insufficient model generalization ability due to single sample sources. The method for constructing an original image library provided in this application focuses on multi-source sample collection, expert identification assurance, standardized imaging, and structured storage to solve the above problems, significantly improving the efficiency of image library construction and the accuracy of intelligent recognition.

[0046] This application will fill the technological gap in the field of intelligent identification of small, medium and large soil animals as well as other small arthropods, provide high-precision data support for biodiversity monitoring and soil health assessment, and promote the transformation of ecological research paradigms towards automation and standardization.

[0047] This application takes medium-sized soil animals as an example to describe in detail the method for constructing the original animal image library provided in the above embodiments, mainly involving the following steps:

[0048] Step 1: Systematic multi-source sample collection and experimental design.

[0049] Images of medium-sized soil animal species are extremely scarce, making it impossible to obtain a sufficient number (e.g., at least 100 images per species) of species images for intelligent identification by downloading from public websites or databases. Therefore, it is necessary to systematically collect samples and specimens of medium-sized soil animals and then obtain a sufficient number of species images for intelligent identification.

[0050] Step 1.1: Before conducting systematic sampling, a detailed work plan and research scheme need to be developed. Traditional systematic sampling of medium-sized soil animals lacks standardized procedures and uniform standards, and a work plan and research scheme need to be developed based on the research objectives, research areas, and research subjects. Unlike traditional systematic sampling of medium-sized soil animals, the work plan and research scheme for systematic sampling of medium-sized soil animals for image intelligent recognition should particularly emphasize the following: (1) Obtaining a rich variety of species and individuals is one of the research objectives, so as to ensure that enough medium-sized soil animal specimens are obtained for image taking and intelligent recognition, ensuring that the number of individual samples of each species is ≥10; (2) Fully consider diverse habitats and multiple plots, including at least 2 typical habitats, so as to increase the heterogeneity of habitat sources and plot sources of medium-sized soil animals as one of the research objectives. This is because it can obtain rich external morphological variability of the same species, so as to ensure that the species images taken have diverse and complex external morphological characteristics of the same species.

[0051] Step 1.2: Conduct field sample collection. Methods for collecting medium-sized soil animal samples include soil column method, trap method, and siphon method. There is no unified method or procedure in this application. Appropriate methods should be adopted according to the research purpose, research area, and research subjects.

[0052] Step 1.3: Perform indoor sample separation to obtain medium-sized soil animal specimens. Medium-sized soil animals were separated from the soil using the Tullgeren dry funnel method and placed in specimen bottles containing 75% ethanol (concentration error ±1%). After labeling, these specimens were refrigerated and stored at -20°C for ≤12 months to prevent morphological damage.

[0053] Step 2: Precise morphological identification led by taxonomic experts.

[0054] Accurate species identification is the foundation and prerequisite for highly precise intelligent identification of medium-sized soil animals. This requires medium-sized soil taxonomists to use morphological identification methods to accurately classify specimens based on morphological characteristics (such as the reproductive plate structure and setae arrangement of beetles). Different medium-sized soil animals have different classification methods and systems; there is no unified standard, method, or procedure. Taxonomists need to classify and identify them according to the morphological classification methods of specific groups to ensure the authority and reliability of species annotation.

[0055] Step 3: Standardized microscopic imaging and multi-angle feature capture.

[0056] Images of medium-sized soil animals were taken using a stereomicroscope, and species annotation was completed for each image. The specific steps involved are as follows:

[0057] Step 3.1: Specimen Selection. Selecting medium-sized soil animal specimens for image acquisition aims to enhance the variability and heterogeneity of external morphological features in the original images of medium-sized soil animals, ensuring their external morphological characteristics are fully represented, thereby improving the ability and accuracy of subsequent intelligent identification and classification. The principles for specimen selection are as follows:

[0058] First, different individuals of the same species used for taking images must come from different habitats. Taking soil beetles collected from 12 altitudinal gradient plots in Tianmu Mountain National Nature Reserve, Zhejiang Province and 12 altitudinal gradient plots in Guanshan National Nature Reserve, Jiangxi Province as an example, specimens of the same beetle species used for photography must be collected from the two mountain habitats of Tianmu Mountain and Guanshan respectively.

[0059] Second, different individuals of the same species used for image capture must come from different sample plots. Taking soil beetles collected from 12 altitudinal gradient sample plots in Tianmu Mountain National Nature Reserve, Zhejiang Province as an example, specimens of the same beetle species used for image capture should cover all altitudinal gradient sample plots of that species. When there are fewer than 5 sample plots of a species, specimens from all sample plots must be selected for image capture.

[0060] Third, medium-sized soil animal specimens used for image capture must be physically intact. Select intact medium-sized soil animal specimens for image capture. Do not select individuals with incomplete or damaged external morphological characteristics for image capture, including those with missing antennae, decaying bodies, missing appendages, dried-out bodies, or bodies that have been compressed or deformed.

[0061] Step 3.2: Specimen Preparation. Prepare a medium-sized soil animal specimen for photographing, ensuring adequate preparation for the photography process. This includes the following steps:

[0062] Step 3.2.1: Take the medium-sized soil animal specimen bottle out of the refrigerator, wipe off the water droplets on the surface of the specimen bottle, and place the specimen bottle near the stereomicroscope.

[0063] Step 3.2.2: Select the substrate for the stereomicroscope for subsequent photography, ensuring the substrate color differs significantly from the characteristics of the medium-sized soil animal specimen. For example, since the body color of the long-winged beetle is dark, a white substrate should be selected.

[0064] Step 3.2.3: Take out a transparent glass petri dish, wipe it clean with a tissue, and then place it near the stereomicroscope, close to the aforementioned medium-sized soil animal specimen bottle.

[0065] Step 3.2.4: Open the medium-sized soil animal specimen bottle, tilt the petri dish at an angle ≤15°, and transfer all the solution from the specimen bottle into the petri dish. Then rinse the specimen bottle several times with a wash bottle containing 75% alcohol until all the medium-sized soil animal samples in the specimen bottle have been transferred to the petri dish. This step requires special care to prevent alcohol from dripping outside the petri dish and causing specimen loss.

[0066] Step 3.2.5: Place the petri dish containing the medium-sized soil animal specimen in the center of the stereomicroscope plate, locate the medium-sized soil animal specimen to be photographed by microscopic examination, select one medium-sized soil animal specimen, and place it into a second clean glass petri dish.

[0067] Step 3.2.6: Remove the first petri dish containing the medium-sized soil animal specimen from the stereomicroscope slide and gently place it near the stereomicroscope. This step requires ensuring the petri dish is stable to prevent the solution from splashing out or spilling, which could result in the loss of the specimen.

[0068] Step 3.2.7: Place the second petri dish containing one medium-sized soil animal specimen in the center of the stereomicroscope for subsequent image taking.

[0069] Step 3.3: Multi-angle image capture. Capture images of medium-sized soil animals from multiple angles, including the dorsal, ventral, and lateral sides, to obtain detailed morphological features of their complex body surfaces, providing high-quality image data for intelligent recognition. Taking the species *Pergalumna intermedia* as an example, the steps include:

[0070] Step 3.3.1: Use soft-tipped tweezers to place the first medium-sized soil animal specimen under the stereomicroscope with its back facing up, and center its image in the computer imaging system. Use a stereomicroscope equipped with a ring LED cold light source to perform shadowless imaging.

[0071] Step 3.3.2: Adjust the objective lens to magnify the first medium-sized soil animal specimen to the maximum magnification (e.g., 40x-110x), with an image resolution of no less than 20 million pixels, ensuring that the entire individual is in the computer imaging system.

[0072] Step 3.3.3: After the first medium-sized soil animal specimen in the computer imaging system has come to a standstill, click the mouse to take an image of the specimen.

[0073] Step 3.3.4: Export the image of the first medium-sized soil animal specimen to a folder in JPG format. Name the folder as “Date_Location_Habitat_Plot_Species Name_Photo Number”. The date is the date the specimen was surveyed, expressed in year and month, e.g., June 2021 is represented as 202106. Location is the location where the specimen was surveyed, e.g., Tianmu Mountain is represented as Zhejiang Tianmu Mountain. Habitat is the habitat type, e.g., forest is represented as forest habitat. Plot is the name or number of the plot where the specimen was surveyed, e.g., TM800 is represented as plot number TM800. Species name is the species name, e.g., Protoribates capucinus is the species name of the beetle. Photo number is the sequence number of the image taken for this species, e.g., 16 is the 16th image taken for this species. The overall format is: 202106_Tianmu Mountain_Forest_TM800_Pergalumna intermedia_16.

[0074] Step 3.3.5: Continue imaging using multi-angle continuous imaging technology combined with specimen dynamic positioning method. Gently change the angle of the first medium-sized soil animal specimen with soft-tipped tweezers, rotating the first medium-sized soil animal specimen clockwise by 30°, 60° or 90° to change the position of the first medium-sized soil animal specimen in the computer imaging system. Then repeat steps 3.3.3 and 3.3.4 to take and save the second back image of the first medium-sized soil animal specimen.

[0075] Step 3.3.6: Repeat step 3.3.5 to take at least 5 images of the back of the first medium-sized soil animal specimen.

[0076] Step 3.3.7: Repeat steps 3.3.1 to 3.3.6 to take at least 5 images of the ventral side of the first medium-sized soil animal specimen.

[0077] Step 3.3.8: Repeat steps 3.3.1 to 3.3.6 to take at least 5 side views of the first medium-sized soil animal specimen.

[0078] Step 3.3.9: After taking the first image, put the specimen back into the original specimen bottle and immerse it in 75% alcohol.

[0079] Step 3.3.10: Repeat steps 3.3.1 to 3.3.9 to take images of the second beetle species, Protoribates capucinus, until all specimens that need to be photographed have been photographed.

[0080] Step 4: Structure the image database architecture.

[0081] After completing the image acquisition for the beetle, check the quantity, quality, and naming of all images to ensure they are correct. Once confirmed, construct the original image database. The database architecture uses a five-level directory structure: "Date → Location → Habitat → Plot → Species." Create new folders according to the format "Date_Location_Habitat_Plot_Species Name," placing images with the same date, location, habitat, plot, and species name in the same folder. For example, images named "202106_Tianmushan_Forest_TM800_Protoribates capucinus_1" to "202106_Tianmushan_Forest_TM800_Protoribates capucinus_n" are stored together in a folder named "202106_Tianmushan_Forest_TM800_Protoribates capucinus".

[0082] Simultaneously, data redundancy control is implemented: duplicate images are filtered out using SSIM similarity analysis (threshold > 95%) to ensure that the image library for each species has ≥ 100 images and morphological feature coverage > 90%. This completes the construction of a primary image library for medium-sized soil animals used for intelligent identification. A highly available image library is constructed based on a five-level directory (date, location, habitat, plot, species) and SSIM similarity deduplication (threshold > 95%).

[0083] The poor quality of existing animal image databases is mainly due to: (1) Data scarcity: Due to the small size of the animals (requiring 40x-110x microscopic imaging) and the specialization of operation, existing open databases (such as Collembola.org) only have 3 to 5 images of a single species, which is far from reaching the threshold for training intelligent models (≥100 images of a single species). This application uses 40x-110x microscopic imaging and a single species image database of ≥100 to solve this problem; (2) Data quality defects: Existing image databases are mostly derived from laboratory specimens and lack multi-habitat sources (such as across climate zones and altitude gradients), resulting in low coverage of morphological variations of the same species. This application obtains specimens from multiple habitats and diverse locations to solve this problem; (3) Lack of standardization: The image acquisition process is highly arbitrary (such as uneven lighting and single angle), and there is no standardized acquisition process. This application solves this problem by acquiring images from multiple rotation angles and using a stereomicroscope equipped with a ring LED cold light source for shadowless imaging.

[0084] This application uses systematic sampling across multiple habitats and diverse locations to ensure that each species covers ≥2 typical habitats and ≥5 sample plots (all plots are selected if the number of sample plots is insufficient), with a single species sample size of ≥100 images. This significantly improves the coverage of morphological variability within the same species and effectively solves the problem of model overfitting caused by the single source of samples in existing technologies.

[0085] This application utilizes precise morphological identification led by taxonomic experts, combined with specimen integrity screening, to ensure a species annotation error rate of <1% (compared to a typical error rate of >10% in traditional public databases), providing highly reliable annotation data for intelligent identification and ensuring the authority and reliability of species annotation.

[0086] This application employs standardized microscopic imaging technology, using 40-110x magnification, 20 million pixel resolution, and multi-angle rotation shooting (30°, 60°, 90°) to capture ≥15 images of a single object (5 images each of the back, ventral, and side views), which improves the completeness of image feature extraction by more than 50% compared to traditional single-angle shooting.

[0087] A stereomicroscope equipped with a ring-shaped LED cold light source enables shadowless imaging, avoiding feature loss caused by uneven illumination.

[0088] By using soft-tipped tweezers for dynamic positioning, the three-dimensional morphological features of the specimen can be covered, thus solving the blind spots in identification caused by the single angle of traditional methods.

[0089] This application is based on a database design with a five-level directory structure (date → location → habitat → plot → species), combined with SSIM similarity analysis (threshold > 95%), to achieve fast image retrieval and efficient management, and improves data retrieval efficiency by 70% compared with traditional flat storage.

[0090] This application, through standardized parameter design (such as 75% alcohol preservation error ±1%, refrigeration time ≤12 months), can be seamlessly transferred to the construction of image libraries of other micro soil animals (such as springtails and nematodes), solving the problem of process reconstruction caused by taxonomic differences in traditional methods.

[0091] This application achieves expanded morphological diversity coverage through systematic multi-source sampling, ensures the authority of data annotation through expert evaluation, enhances feature integrity through high-precision and multi-angle imaging, and improves data usability through structured databases. These four core technological breakthroughs solve the core pain points of low image data quality and poor model generalization ability in the intelligent identification of small, medium and large soil animals and other small arthropods, providing a highly reliable data foundation for animal diversity research and ecological monitoring.

[0092] Based on the same inventive concept, this application also provides an apparatus for constructing an animal original image library to implement the above-described method for constructing an animal original image library. The solution provided by this apparatus is similar to the solution described in the above-described method. Therefore, the specific limitations of one or more embodiments of the animal original image library construction apparatus provided below can be found in the limitations of the animal original image library construction method described above, and will not be repeated here.

[0093] In one exemplary embodiment, such as Figure 2 As shown, an apparatus for constructing an animal original image library is provided, comprising:

[0094] The specimen acquisition module is used to acquire the specimen set corresponding to each species in the target group; the specimen set corresponding to each species includes specimens of species collected from various locations in various habitats; the target group is medium-sized soil animals, small soil animals, large soil animals or arthropods.

[0095] The image acquisition module is used to acquire images of the back, ventral, and lateral sides of any specimen at various rotation angles, thereby obtaining the corresponding image library for the specimen.

[0096] The image library construction module is used to construct an original animal image library from the image libraries corresponding to all specimens in the specimen collections of each species in the target group.

[0097] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 3 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data for constructing an animal image database. The I / O interfaces allow the processor to exchange information with external devices. The communication interface allows communication with external terminals via a network connection. When executed by the processor, the computer program implements a method for constructing an animal image database.

[0098] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0099] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described method embodiments.

[0100] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the above-described method embodiments.

[0101] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the above-described method embodiments.

[0102] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0103] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0104] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0105] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible 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.

[0106] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for constructing a raw animal image database, characterized in that, The method for constructing the original animal image library includes: Obtain the specimen collection corresponding to each species in the target group; the specimen collection corresponding to each species includes specimens of species collected from various locations in various habitats; the target group is a medium-sized soil animal, a small soil animal, a large soil animal, or an arthropod. For any specimen, images of the back, ventral, and lateral sides of the specimen are acquired at various rotation angles to obtain the corresponding image library for the specimen; The original animal image database is constructed by creating an image library of all specimens in the specimen collections corresponding to each species in the target group.

2. The method for constructing an animal original image library according to claim 1, characterized in that, For any specimen, images of the dorsal, ventral, and lateral surfaces of the specimen are acquired at various rotation angles to obtain a corresponding image library, specifically including: Using a stereomicroscope equipped with a ring-shaped LED cold light source, when the entire specimen is located in the computer imaging system, the back, ventral and lateral surfaces of the specimen are imaged without shadows at various rotation angles, resulting in images of the back, ventral and lateral surfaces of the specimen at various rotation angles. Images of the dorsal, ventral, and lateral surfaces of the specimen at various rotation angles are named according to the specimen parameters and image number to obtain the corresponding image library for the specimen. The specimen parameters include the specimen acquisition time, specimen acquisition location, specimen acquisition habitat, specimen acquisition plot, and the species name corresponding to the specimen.

3. The method for constructing the original animal image library according to claim 2, wherein the image library corresponding to all specimens in the specimen set corresponding to each species in the target group is constructed as the original animal image library, specifically including: Store images of specimens with identical parameters in the same folder; For any folder, calculate the structural similarity between every two images in the folder; If the structural similarity between two images is greater than a set threshold, then either of the two images will be deleted to obtain the folder after removing redundancy. All folders, after removing redundancy, will be used to build a library of original animal images.

4. The method for constructing an animal original image library according to claim 1, characterized in that, The rotation angles are 30 degrees, 60 degrees and 90 degrees respectively.

5. The method for constructing an animal original image library according to claim 2, characterized in that, The image resolution of a stereo microscope is greater than or equal to 20 million pixels, and 40≤r≤110, where r is the magnification of the stereo microscope.

6. The method for constructing an animal original image library according to claim 1, characterized in that, Obtain the specimen collections corresponding to each species in the target taxa, specifically including: For any given species, animal samples corresponding to that species are collected from various locations in different habitats; The collected animal samples were placed in specimen bottles containing 75% alcohol to obtain a collection of specimens corresponding to the species.

7. An apparatus for constructing an original animal image database, characterized in that, The apparatus for constructing the original animal image library includes: The specimen acquisition module is used to acquire the specimen set corresponding to each species in the target group; the specimen set corresponding to each species includes specimens of the species collected from various locations in various habitats; the target group is medium-sized soil animals, small soil animals, large soil animals, or arthropods; The image acquisition module is used to acquire images of the back, ventral and lateral sides of any specimen at various rotation angles, and obtain the image library corresponding to the specimen. The image library construction module is used to construct an original animal image library from the image libraries corresponding to all specimens in the specimen collections of each species in the target group.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method for constructing the animal original image library as described in claim 1.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for constructing the animal original image library as described in claim 1.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the method for constructing the animal original image library as described in claim 1.

Images