Biological standard fossil database online query method and system

Abstract

The invention discloses a biological standard fossil database online query method and system, and belongs to the technical field of biological fossil recognition processing. The method comprises the following steps: performing illumination influence reduction processing on a to-be-identified fossil image according to light condition data of the to-be-identified fossil image and glossiness of the to-be-identified fossil; performing crack influence reduction processing on a crack area of the to-be-identified fossil image after illumination influence reduction processing according to the crack intensity parameter of the to-be-identified fossil image after illumination influence reduction processing; similarity evaluation is performed on the to-be-identified fossil image subjected to crack influence reduction processing and each biological standard fossil image in the biological standard fossil database, and the to-be-identified fossil type is divided according to a similarity evaluation result, so that interference of cracks on image analysis and feature extraction is reduced; therefore, the image features can truly reflect the structure of the fossil, and the accuracy of the query result is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of biological fossil identification and processing, and in particular to an online query method and system for a biological standard fossil database. Background Art

[0002] The online query method of the biological standard fossil database is a digital tool established to conveniently access and analyze fossil data. By collecting and organizing a large amount of fossil specimen information, it provides important support for research in paleontology, geology and other fields.

[0003] The existing online query method of the biological standard fossil database builds a centralized digital platform, adopts a standardized data format and a unified identification system, and systematically stores and manages the relevant information of fossil specimens. Users can quickly retrieve the required fossil data through keyword search, species classification, geographic location, stratigraphic age and other query methods, and use modern database technology and network interfaces to ensure the accuracy, accessibility and efficiency of the data.

[0004] For example, the invention patent with announcement number: CN113128335B announces the method, system and application of detecting, classifying and discovering images of microfossils, including: formulating microfossil image acquisition standards and taking images of microfossils; constructing a data set with simulated microfossils; building an SSD network; adjusting the aspect ratio of the pre-selected box; loading the original weight file of the pre-trained model and training the network model for microfossil image detection; inputting the image to be detected into the trained network model and using the non-maximum suppression algorithm to screen out suitable detection results; and making special records of the detection results of artificial simulated fossil categories that are different from the original known microfossils.

[0005] For example, a patent application with publication number: CN116524243A discloses a method and device for classifying graptolite fossil images, which includes: preparing a graptolite fossil image dataset; building and training a metadata embedding classification model, which includes an embedding model and a classification layer model; building and training a graptolite fossil single image classifier; inputting the graptolite fossil single image and the corresponding metadata in the test set into the graptolite fossil single image classifier and the metadata embedding classification model respectively, and fusing the outputs of the two to obtain the graptolite fossil single image category prediction result.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0007] In the prior art, when querying the species of a fossil to be identified using a biological standardized fossil database, since the fossil to be identified may have cracks, which affects the surface state of the fossil to be identified, the image quality of the fossil to be identified during scanning or photographing decreases, and effective morphological information cannot be accurately identified and extracted. There is a problem of low accuracy in querying the fossil to be identified using the biological standardized fossil database. Summary of the Invention

[0008] The present invention provides an online query method and system for a biological standardized fossil database, which solves the problem in the prior art that when querying the species of a fossil to be identified using a biological standardized fossil database, since the fossil to be identified may have cracks, which affects the surface state of the fossil to be identified, the image quality of the fossil to be identified during scanning or photographing decreases, and effective morphological information cannot be accurately identified and extracted. There is a problem of low accuracy in querying the fossil to be identified using the biological standardized fossil database, and achieves the goal of improving the accuracy of fossil species query.

[0009] The present invention provides an online query method for a biological standardized fossil database, including the following steps: dividing the image of the fossil to be identified into a crack area and a non-crack area, reducing the influence of illumination on the image of the fossil to be identified according to the light condition data of the image of the fossil to be identified and the glossiness of the fossil to be identified, and the illumination influence reduction process is used to reduce the influence of illumination difference on the accuracy of feature extraction of the image of the fossil to be identified; reducing the influence of cracks on the image of the fossil to be identified according to the crack intensity parameter of the image of the fossil to be identified after the illumination influence reduction process for the crack area of the image of the fossil to be identified after the illumination influence reduction process, and the crack influence reduction process is used to reduce the influence of crack difference on the similarity evaluation of the image of the fossil to be identified; evaluating the similarity between the image of the fossil to be identified after the crack influence reduction process and each biological standardized fossil image in the biological standardized fossil database, and dividing the species of the fossil to be identified according to the similarity evaluation result.

[0010] The present invention provides an online query system for a biological standardized fossil database, including a light influence reduction processing module, a crack influence reduction processing module, an online query module, and a fossil query database; wherein, the light influence reduction processing module is used to divide the fossil image to be identified into a crack area and a non-crack area, and perform light influence reduction processing on the fossil image to be identified according to the light condition data of the fossil image to be identified and the glossiness of the fossil to be identified. The light influence reduction processing is used to reduce the influence of light difference on the accuracy of feature extraction of the fossil image to be identified; the crack influence reduction processing module is used to perform crack influence reduction processing on the crack area of the fossil image to be identified after the light influence reduction processing according to the crack intensity parameter of the fossil image to be identified after the light influence reduction processing. The crack influence reduction processing is used to reduce the influence of crack difference on the similarity evaluation of the fossil image to be identified; the online query module is used to perform similarity evaluation on the fossil image to be identified after the crack influence reduction processing and each biological standardized fossil image in the biological standardized fossil database, and divide the types of fossils to be identified according to the similarity evaluation result.

[0011] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0012] 1. By providing a method and system for querying a biological standardized fossil database, the present invention performs light influence reduction processing on the fossil image to be identified. Based on the crack intensity parameter after the light influence reduction, the interference of the crack area is further reduced. By performing similarity evaluation with the biological standardized fossil images in the biological standardized fossil database, the types of fossils to be identified can be divided more accurately, thereby realizing an efficient and more accurate fossil query and identification process.

[0013] 2. By adaptively performing light influence reduction processing according to the light condition of the image, the present invention reduces the influence of light according to different deviation light disturbance evaluation indexes, enhances the image details, and thereby realizes the precise optimization processing of the fossil image according to different light conditions, improving the image quality in the fossil identification process.

[0014] 3. By quantifying the influence of cracks on the fossil image to be identified after the light influence reduction processing according to the crack intensity parameter, obtaining the crack intensity index, and comparing it with a preset threshold, the present invention adopts different filling and repair methods according to the range of different crack intensity indexes, thereby realizing the precise repair of the crack area, improving the image quality, making the identification of fossils more accurate, and reducing the situation where the identification cannot be carried out due to excessive cracks.

[0015] 4. The present invention extracts the edge contour, surface texture, and high-dimensional feature vectors of the fossil image to be identified after reducing the influence of cracks, and evaluates the similarity with the images in the biological standard fossil database, thereby quantifying the similarity between the fossil to be identified and each biological standard fossil image. Furthermore, by comparing the maximum correlation scoring coefficient with a preset threshold, the type of the fossil to be identified is accurately determined, avoiding misjudgment and improving the accuracy of fossil identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of a method for online querying of a biological standard fossil database provided by an embodiment of the present application.

[0017] Figure 2 It is a graph showing the change of the crack strength index of a method for online querying of a biological standard fossil database provided by an embodiment of the present application.

[0018] Figure 3 It is a schematic structural diagram of a system for online querying of a biological standard fossil database provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] In an embodiment of the present application, by providing a method and system for online querying of a biological standard fossil database, the problem in the prior art that when querying the type of a fossil to be identified using a biological standard fossil database, the surface state of the fossil to be identified may be affected due to possible cracks in the fossil to be identified, resulting in a decrease in the quality of the scanned or photographed image of the fossil to be identified, and it is impossible to accurately identify and extract effective morphological information, and there is a problem of low accuracy in querying the fossil to be identified using a biological standard fossil database is solved. The image of the fossil to be identified is divided into a crack area and a non-crack area, and the influence of illumination on the image of the fossil to be identified is reduced according to the light condition data and the glossiness of the fossil to be identified of the image of the fossil to be identified. The illumination influence reduction process is used to reduce the influence of illumination difference on the accuracy of feature extraction of the image of the fossil to be identified; according to the crack strength parameter of the image of the fossil to be identified after the illumination influence reduction process, the crack area of the image of the fossil to be identified after the illumination influence reduction process is subjected to a crack influence reduction process. The crack influence reduction process is used to reduce the influence of crack difference on the similarity evaluation of the image of the fossil to be identified; the image of the fossil to be identified after the crack influence reduction process is evaluated for similarity with each biological standard fossil image in the biological standard fossil database, and the type of the fossil to be identified is divided according to the similarity evaluation result, achieving the goal of improving the accuracy of querying the type of fossil.

[0020] The technical solution in the embodiment of this application aims to solve the problem that when using a biological standard fossil database to query the type of the fossil to be identified, the surface state of the fossil to be identified may be affected by cracks, resulting in a decrease in the image quality of the fossil to be identified during scanning or photographing, and the inability to accurately identify and extract effective morphological information, leading to a low accuracy in querying the fossil to be identified using the biological standard fossil database. The general idea is as follows:

[0021] By processing the image based on the light condition data and the fossil glossiness to reduce the influence of illumination, the interference of illumination differences on the extraction of fossil image features is reduced; by processing the crack area to reduce the influence of cracks, the interference of crack differences on the similarity evaluation is reduced; by evaluating the similarity between the image after reducing the influence of cracks and each standardized fossil image in the biological standard fossil database, the similarity between the fossil to be identified and the fossil images in the database can be accurately compared. According to the similarity evaluation results, the types of the fossils to be identified can be effectively classified, achieving an improvement in the accuracy and automation level of fossil identification.

[0022] To better understand the above technical solution, the above technical solution will be described in detail below in combination with the accompanying drawings of the specification and specific implementation manners.

[0023] As Figure 1 shown, it is a flowchart of a method for online querying of a biological standard fossil database provided by an embodiment of this application. The method includes the following steps: dividing the image of the fossil to be identified into a crack area and a non-crack area, processing the image of the fossil to be identified based on the light condition data and the glossiness of the fossil to be identified to reduce the influence of illumination, and the illumination influence reduction processing is used to reduce the influence of illumination differences on the accuracy of feature extraction of the image of the fossil to be identified. The light condition data includes the proportion of high-light pixels, the standard gray difference, and the shadow coverage rate. The proportion of high-light pixels represents the pixel ratio of each pixel point in the image of the fossil to be identified with an RGB value greater than the preset RGB threshold. The standard gray difference represents the standard deviation of the gray values of each pixel in the image of the fossil to be identified from the average gray value. The shadow coverage rate represents the proportion of the shadow area in the total area of the image of the fossil to be identified; processing the crack area of the image of the fossil to be identified after the illumination influence reduction processing based on the crack intensity parameter of the image of the fossil to be identified after the illumination influence reduction processing, and the crack influence reduction processing is used to reduce the influence of crack differences on the similarity evaluation of the image of the fossil to be identified; evaluating the similarity between the image of the fossil to be identified after the crack influence reduction processing and each biological standard fossil image in the biological standard fossil database, and classifying the types of the fossils to be identified according to the similarity evaluation results.

[0024] In this embodiment, the present invention is applied to the processing of the fossil image to be identified before online query of the biological standardized fossil database. By separately processing the crack area and the non-crack area in the image, the influence of cracks on the image quality can be more accurately identified, and targeted repair can be carried out in subsequent processing, reducing the interference of cracks on the extraction of the entire image features, thereby improving the accuracy of morphological information extraction; by reducing the influence of illumination on the image according to the light conditions and glossiness, the influence brought by illumination differences can be eliminated or reduced, ensuring that the feature extraction of the image is still accurate under different illumination conditions, thereby improving the quality of the fossil image to be identified; the cracks of the fossil to be identified may affect the surface features, making the traditional image similarity evaluation method no longer effective. By specially reducing the influence of cracks on the crack area, the influence of cracks on the image matching process can be reduced.

[0025] In addition, the fossil query database is used to store relevant data based on the online query method of the biological standardized fossil database, including: the illumination disturbance evaluation threshold corresponding to each glossiness value, the highlight pixel influence factor, the critical highlight pixel ratio, the critical standard gray difference, and the critical shadow coverage rate, etc. The data in the fossil query database can be obtained through cooperation with scientific research organizations such as the Palaeontological Society of China, or can be queried through public databases such as the Palaeontological Site Protection Database.

[0026] Further, the steps of reducing the influence of illumination on the fossil image to be identified according to the light condition data and glossiness of the fossil image to be identified include:

[0027] First, match the glossiness of the fossil to be identified with the illumination disturbance evaluation threshold corresponding to each glossiness value in the preset fossil query database to obtain the illumination disturbance evaluation threshold corresponding to the glossiness of the fossil to be identified.

[0028] Then, quantify the influence of light on the quality of the fossil image to be identified according to the light condition data of the fossil image to be identified to obtain the illumination disturbance evaluation index. The specific process is as follows: Obtain the light condition data reference data from the preset fossil query database, specifically including: the critical highlight pixel ratio, the critical standard gray difference, and the critical shadow coverage rate; perform the ratio approximation operation on the highlight pixel ratio, the standard gray difference, and the shadow coverage rate with the critical highlight pixel ratio, the critical standard gray difference, and the critical shadow coverage rate respectively, and then perform the weighted coupling process on the ratio approximation operation results through the light condition data compensation values respectively. The light condition data compensation values include the highlight pixel influence factor, the standard gray difference influence factor, and the shadow coverage rate influence factor. The illumination disturbance evaluation index represents the quantification data of the combined influence degree of the highlight pixel ratio, the standard gray difference, and the shadow coverage rate on the quality of the fossil image to be identified. The acquisition method of the illumination disturbance evaluation index is as follows:

[0029]

[0030] In the formula, LI represents the light disturbance evaluation index, μ1 represents the high-light pixel influence factor, μ2 represents the standard gray difference influence factor, μ3 represents the shadow coverage rate influence factor, HP1 represents the proportion of high-light pixels, HP0 represents the proportion of critical high-light pixels, GD1 represents the standard gray difference, GD0 represents the critical standard gray difference, SC1 represents the shadow coverage rate, and SC0 represents the critical shadow coverage rate.

[0031] μ1, μ2, and μ3 are respectively the compensation values corresponding to the preset proportion of high-light pixels, standard gray difference, and shadow coverage rate in the fossil query database, and respectively represent the numerical values of the influence degrees of the proportion of high-light pixels, standard gray difference, and shadow coverage rate on the light disturbance evaluation index, and can be directly obtained from the fossil query database when used. For example, the proportion of high-light pixels forms a mapping set with the compensation value corresponding to the preset proportion of high-light pixels in the fossil query database, and the compensation value corresponding to the proportion of high-light pixels is obtained by inputting the proportion of high-light pixels into the mapping set; the standard gray difference forms a mapping set with the compensation value corresponding to the preset standard gray difference in the fossil query database, and the compensation value corresponding to the standard gray difference is obtained by inputting the standard gray difference into the mapping set; the shadow coverage rate forms a mapping set with the compensation value corresponding to the preset shadow coverage rate in the fossil query database, and the compensation value corresponding to the shadow coverage rate is obtained by inputting the shadow coverage rate into the mapping set, where the mapping relationship is a many-to-one or one-to-one relationship, and in this embodiment, the value range of the compensation value is between 0 and 1.

[0032] Finally, compare the light disturbance evaluation index with the light disturbance evaluation threshold. If the light disturbance evaluation index is less than or equal to the light disturbance evaluation threshold, no additional operation is performed. Otherwise, the fossil image to be identified is processed to reduce the light influence according to the deviation light disturbance evaluation index, and the deviation light disturbance evaluation index represents the deviation between the light disturbance evaluation index and the light disturbance evaluation threshold.

[0033] In this embodiment, the glossiness value can be measured by using a glossmeter to measure the surface of the fossil. The instrument emits light at multiple specific angles (such as 20°, 60°, and 85°), receives the reflected light, and obtains the average value of the reflected intensity through calculation. Among them, the light perturbation evaluation threshold represents the critical point at which the light source conditions have a significant impact on the glossiness of the fossil, and it can be directly obtained from the fossil query database. For example, in the fossil query database, a mapping relationship table is formed by corresponding the glossiness value with the light perturbation evaluation threshold one by one. The table records each glossiness value and its corresponding light perturbation evaluation threshold. These relationships can be one-to-one or many-to-one. When obtaining the light perturbation evaluation threshold, only need to input the glossiness value into the mapping relationship table, and the fossil query database can quickly locate and return the light perturbation evaluation threshold corresponding to this glossiness value. The light influence reduction process can retain more details of the fossil surface by reducing the over-strong or uneven light reflection. Through the light influence reduction process, the negative impact of light on the image quality can be reduced, highlighting the true features of the fossil, and ensuring that the morphology and features of the fossil can be accurately analyzed during subsequent identification.

[0034] In addition, the proportion of high-light pixels, the standard gray difference, and the shadow coverage rate are interrelated. For example, the larger the standard gray difference, the higher the shadow coverage rate may be; if the proportion of the high-light area in the image is larger and the standard gray difference is large, there may be a strong light influence in the image, which may lead to the loss of fossil details or overexposure. By combining these three indicators, the light influence of the fossil image to be identified can be evaluated more comprehensively. The light perturbation evaluation index obtained through comprehensive analysis can accurately obtain the degree of influence of the image in terms of light, so as to evaluate the quality of the fossil image to be identified.

[0035] Further, the steps of reducing the illumination impact on the fossil image to be identified according to the deviation illumination disturbance evaluation index include: obtaining a first deviation illumination disturbance evaluation threshold and a second deviation illumination disturbance evaluation threshold from a preset fossil query database; comparing the deviation illumination disturbance evaluation index with the first deviation illumination disturbance evaluation threshold and the second deviation illumination disturbance evaluation threshold respectively. If the deviation illumination disturbance evaluation index is less than the first deviation illumination disturbance evaluation threshold, restricted histogram equalization is enabled, and at the same time, the contrast of the crack area is enhanced according to the deviation illumination disturbance evaluation index. Specifically, a threshold is preset to limit the number of pixels of a single gray level in the histogram, and the excess part is truncated and evenly distributed to other gray levels. Then the image is divided into several small blocks, and histogram equalization is performed independently on each small block, and the Retinex shadow enhancement technology is used to enhance the shadow area. If the deviation illumination disturbance evaluation index is greater than or equal to the first deviation illumination disturbance evaluation threshold and less than the second deviation illumination disturbance evaluation threshold, the non-subsampled contourlet transform technology is used to decompose the fossil image to be identified into high-frequency components and low-frequency components, homomorphic filtering is applied to the low-frequency components, and adaptive threshold denoising is performed on the high-frequency components. If the deviation illumination disturbance evaluation index is greater than or equal to the second deviation illumination disturbance evaluation threshold, the multi-scale Gaussian surround is used to separate the illumination and reflection components, the reflection component is eliminated from the fossil image to be identified, and at the same time, adaptive homomorphic filtering is enabled (dynamically adjusting the cut-off frequency and gain for the V channel in the HSV space, suppressing the low-frequency illumination while retaining the high-frequency cracks).

[0036] Among them, the steps of enhancing the contrast of the crack area according to the deviation illumination disturbance evaluation index include: obtaining the contrast enhancement value corresponding to each deviation light influence area from a preset fossil query database; matching the deviation illumination disturbance evaluation index with each deviation light influence area to obtain the deviation light influence areas corresponding to the deviation illumination disturbance evaluation index, so as to obtain the contrast enhancement value corresponding to the deviation illumination disturbance evaluation index; processing the contrast of the fossil image to be identified according to the contrast enhancement value; the contrast enhancement values include 10, 15, and 20.

[0037] In this embodiment, by obtaining and comparing the deviation light disturbance evaluation index with a preset threshold, the system can perform adaptive processing according to the specific light interference situation of the image, avoiding a one-size-fits-all processing method, thus retaining more details and reducing ineffective or excessive correction. By enabling restricted histogram equalization processing, the brightness and contrast of the image can be effectively improved, and the details on the fossil surface, especially important features such as cracks, can be enhanced. The processing of the low-frequency components by homomorphic filtering helps to reduce the brightness variation caused by uneven ambient light, making the illumination of the fossil image more uniform. The adaptive denoising of the high-frequency components can effectively remove noise while retaining the detailed structure in the image, making the details of the fossil more prominent. For images severely affected by reflection interference, the system will separate the light and reflection components in the image and eliminate the reflection component, which helps to remove the interference brought by gloss and specular reflection, making the true surface structure of the fossil clearer. Adaptive homomorphic filtering helps to dynamically adjust the image under different lighting conditions, especially suitable for images with more prominent reflection components. Adaptive filtering can flexibly adjust the filtering intensity according to the changes in the image content and light distribution, thus maintaining the details and texture of the image.

[0038] The contrast enhancement value reflects the degree of enhancement of the image contrast and can be directly obtained from the fossil query database. For example, in the fossil query database, a mapping relationship table is formed by corresponding each deviation light influence area with the contrast enhancement value one by one. The table records each deviation light influence area and its corresponding contrast enhancement value, and these relationships are one-to-one. When obtaining the contrast enhancement value, only the deviation light disturbance evaluation index needs to be input into the mapping relationship table, and the fossil query database can quickly locate the deviation light influence area corresponding to the deviation light disturbance evaluation index and return the contrast enhancement value corresponding to the deviation light influence area. The setting of the contrast enhancement value can flexibly cope with different lighting conditions, ensuring that the crack and important feature areas can still be clearly presented under the influence of light deviation. This targeted enhancement ensures the consistency and reliability of the image in different lighting environments. When processing the fossil image to be identified according to the contrast enhancement value, the value obtained by directly adding the image contrast and the contrast enhancement value can be used as the contrast value of the current image.

[0039] Further, the steps of reducing the impact of cracks on the cracked area of the fossil image to be identified after the reduction of the impact of light include: First, quantify the impact of cracks on the fossil image to be identified after the reduction of the impact of light according to the crack strength parameters to obtain a crack strength index. The crack strength parameters include the proportion of crack area, the proportion of crack depth, and the number of cracks. The proportion of crack area represents the ratio of the total area of the cracked area in the image to the total area of the image. The proportion of crack depth represents the ratio of the maximum depth of the crack to the maximum depth of the fossil to be detected. The number of cracks represents the total number of cracks identified in the image. The parameters are interrelated. For example, if the proportion of crack area is large and the number of cracks is large, it usually means that the surface of the object is severely damaged and dispersed, and the degree of damage is complex. If the proportion of crack area is large and the proportion of crack depth is also high, it usually indicates that the object is severely damaged, with both large-area cracks and deep cracks, which may affect the structural stability of the object. The comprehensively analyzed crack strength index can reflect the damage situation of the object to be identified, evaluate the severity of the cracks and the integrity of the object, so as to judge the impact of the distribution, depth, and number of cracks on the fossil image to be identified.

[0040] The method for obtaining the crack strength index is as follows: Obtain the reference data of crack strength parameters from a preset fossil query database, specifically including: the critical proportion of crack area, the critical proportion of crack depth, and the critical number of cracks; perform a ratio approximation operation on the proportion of crack area, the proportion of crack depth, and the number of cracks respectively with the critical proportion of crack area, the critical proportion of crack depth, and the critical number of cracks, and then perform a weighted coupling process on the ratio approximation operation results through the crack strength parameter compensation values respectively. The crack strength parameter compensation values include the crack area proportion impact factor, the crack depth proportion impact factor, and the number of cracks impact factor. The crack strength index represents the quantitative data of the combined impact degree of the proportion of crack area, the proportion of crack depth, and the number of cracks on the cracked area of the fossil image to be identified after the reduction of the impact of light.

[0041] The way to obtain the crack strength index is as follows:

[0042]

[0043] In the formula, CS represents the crack strength index, μ4 represents the crack area proportion impact factor, μ5 represents the crack depth proportion impact factor, μ6 represents the number of cracks impact factor, CN1 represents the number of cracks, CA1 represents the proportion of crack area, CA0 represents the critical proportion of crack area, CD1 represents the proportion of crack depth, CD0 represents the critical proportion of crack depth, CN1 represents the number of cracks, and CN0 represents the critical number of cracks.

[0044] μ4, μ5, and μ6 are the compensation values corresponding to the preset crack area ratio, crack depth ratio, and number of cracks in the fossil query database respectively, representing the numerical values of the influence degrees of the crack area ratio, crack depth ratio, and number of cracks on the crack strength index, and can be directly obtained from the fossil query database during use. For example, the crack area ratio forms a mapping set with the compensation value corresponding to the preset crack area ratio in the fossil query database, and the compensation value corresponding to the crack area ratio is obtained by inputting the crack area ratio into the mapping set; the crack depth ratio forms a mapping set with the compensation value corresponding to the preset crack depth ratio in the fossil query database, and the compensation value corresponding to the crack depth ratio is obtained by inputting the crack depth ratio into the mapping set; the number of cracks forms a mapping set with the compensation value corresponding to the preset number of cracks in the fossil query database, and the compensation value corresponding to the number of cracks is obtained by inputting the number of cracks into the mapping set, where the mapping relationship is a many-to-one or one-to-one relationship, and the value range of the compensation value in this embodiment is between 0 and 1.

[0045] Second, obtain the first threshold of the crack strength index and the second threshold of the crack strength index from the preset fossil query database; compare the crack strength index with the first threshold of the crack strength index and the second threshold of the crack strength index respectively. If the crack strength index is less than the first threshold of the crack strength index, fill and repair the crack area of the fossil image to be identified after reducing the influence of light by adjacent pixels in each crack area. If the crack strength index is greater than or equal to the first threshold of the crack strength index and less than the second threshold of the crack strength index, extract texture features and pixel information from the surrounding area of the crack area to fill and repair the crack area of the fossil image to be identified after reducing the influence of light. If the crack strength index is greater than or equal to the second threshold of the crack strength index, it is prompted that the cracks in the fossil to be identified need to be repaired and cannot be recognized.

[0046] In this embodiment, by comparing the crack strength index with a preset first threshold and a second threshold, classification is performed according to the severity of the crack. For example, a crack strength index less than the first threshold of the crack strength index may correspond to a slight crack, a crack strength index greater than or equal to the first threshold but less than the second threshold corresponds to a medium crack, and a crack strength index greater than or equal to the second threshold corresponds to a severe crack, which can guide the selection of subsequent repair treatments. For slight cracks, the method fills and repairs according to the adjacent pixel information around the crack area. By methods such as the mean value or weighted average of neighborhood pixels, the area of the slight crack is repaired to make it blend naturally with the surrounding area as much as possible and reduce visual disharmony. For medium cracks, a texture synthesis method is used to extract texture features from around the crack area and apply them to the crack area to restore the natural texture of the area. For cracks, a non-local mean algorithm is used to calculate the pixel information of similar areas in the image to fill the crack area, and at the same time, a partial differential equation repair method is used to protect the edge information of the crack area. For severe cracks, the method will prompt that the crack cannot be effectively repaired and suggest repairs. This hierarchical repair method can adjust the processing method according to the severity of the crack, so as to achieve a more refined and natural repair effect.

[0047] Set the crack area proportion influence factor to 0.4, the crack depth proportion influence factor to 0.4, the crack number influence factor to 0.2, the crack area proportion to 30%, the critical crack area proportion to 20%, the crack depth proportion to 40%, the critical crack depth proportion to 20%, and the critical crack number to 3. When the crack number is continuously increasing, calculate the crack strength index. As shown in Table 1, the data table of the crack strength index of an online query method for a biological standard fossil database.

[0048] Table 1 Data table of the crack strength index of an online query method for a biological standard fossil database

[0049] Number <![CDATA[CN1 (piece)]]> CS 1 1 1.467 2 2 1.533 3 3 1.6 4 4 1.667 5 5 1.733

[0050] As Figure 2 shown, it is a change diagram of the crack strength index of an online query method for a biological standard fossil database provided by an embodiment of the present application. As shown in Table 1 and Figure 2 it can be seen that when the crack area proportion influence factor, the crack depth proportion influence factor, the crack number influence factor, the crack area proportion, the critical crack area proportion, the crack depth proportion, the critical crack depth proportion, and the critical crack number remain unchanged and the crack number is continuously increasing, the crack strength index also continuously increases.

[0051] Furthermore, the steps of evaluating the similarity between the fossil image to be identified after reducing the crack influence and each biological standard fossil image in the biological standard fossil database and classifying the types of fossils to be identified according to the similarity evaluation results include:

[0052] First, extract the image features of the fossil to be identified after the crack influence reduction process and the image features of each biostandard fossil. The image features include the fossil edge contour point set, the fossil surface texture, and the high-dimensional feature vector. Obtain the edge contour correlation scoring coefficient based on the fossil edge contour point set. The specific process is as follows: Use the Canny edge detection technique to extract the contour point set, calculate the Hu moment vectors respectively, and then calculate using the cosine similarity. Among them, the central moment calculation formula is: In the formula, μ p,q represents the central moment of the image, I(x, y) is the pixel value of the image at the point (x, y), and are the central positions of the image. After normalizing the central moment, combine the normalized second-order and third-order central moments to obtain seven Hu invariant moments. The calculation formula of the cosine similarity is: In the formula, A and B are the Hu moment feature vectors of two images, A·B is the dot product of vector A and vector B, and ‖A‖ and ‖B‖ are the norms of vector A and B respectively. Obtain the texture correlation scoring coefficient based on the fossil surface texture. The specific process is as follows: Use the local binary pattern to extract the fossil surface texture features, and then calculate using the histogram intersection. Among them, the calculation formula of the local binary pattern (LBP) is: In the formula, P c is the gray value of the central pixel, P a is the gray value of the i-th neighborhood pixel, and s(x) is the sign function, defined as The LBP histogram calculation formula is: Among them, H(j) is the frequency of the LBP value of j in the histogram, ‖LBP(x, y) = j is the indicator function, which is 1 when LBP(x, y) = j, otherwise 0, and (x, y) is the pixel coordinate in the image. The calculation formula of the histogram intersection is: Where M is the total number of LBP values, and H1(i) and H2(i) are the i-th elements of the LBP histograms of the two images respectively; the high-dimensional feature correlation scoring coefficient is obtained from the high-dimensional feature vector, and the specific process is as follows: Use a pre-trained CNN model to extract the high-dimensional feature vector, and then calculate it using cosine similarity. The three are interrelated. For example, edges reflect the macroscopic shape and contour of an object, while texture reflects the microscopic structure of the object's surface. The combination of the two can provide more accurate object structure and surface information; edge features provide the basic contour of the image, while high-dimensional features comprehensively consider more complex abstract features; texture features reflect the microscopic surface information of the image, while high-dimensional features contain the deeper feature expressions of this microscopic information. The comprehensively analyzed correlation scoring coefficient can provide an all-round similarity assessment for the biostandard fossil image and the fossil image to be identified, reflecting the similarity degree of the images at the levels of shape, surface texture, and high-level features.

[0053] Then, quantify the similarity between the fossil image to be identified after crack influence reduction processing and each biostandard fossil image in the biostandard fossil database according to the edge contour correlation scoring coefficient, texture correlation scoring coefficient, and high-dimensional feature correlation scoring coefficient, and obtain the correlation scoring coefficient of each biostandard fossil image. The specific process is as follows: Use the similarity compensation value to perform weighted coupling processing on the edge contour correlation scoring coefficient, texture correlation scoring coefficient, and high-dimensional feature correlation scoring coefficient of each biostandard fossil image respectively to obtain the correlation scoring coefficient of each biostandard fossil image. The similarity compensation value includes the edge contour similarity influence factor, texture similarity influence factor, and high-dimensional feature similarity influence factor. The correlation scoring coefficient of each biostandard fossil image represents the quantification data of the combined influence degree of the edge contour correlation scoring coefficient, texture correlation scoring coefficient, and high-dimensional feature correlation scoring coefficient on the similarity degree between the fossil to be identified and each biostandard fossil. The acquisition method of the correlation scoring coefficient of each biostandard fossil image is as follows:

[0054] SE i = μ7×EC 1i + μ8×Te 1i + μ9×HF 1i ;

[0055] Where SE i represents the correlation scoring coefficient of the i-th biostandard fossil image, μ7 represents the edge contour similarity influence factor, μ8 represents the texture similarity influence factor, μ9 represents the high-dimensional feature similarity influence factor, EC1 represents the edge contour correlation scoring coefficient, Te 1i represents the texture correlation scoring coefficient, HF 1iIt represents the high-dimensional feature correlation scoring coefficient, where i is the number of each bio-standardized fossil image, i = 1, 2, 3,..., N, and N is the total number of bio-standardized fossil images.

[0056] μ7, μ8, and μ9 are the compensation values corresponding to the preset edge contour similarity, texture similarity, and high-dimensional feature similarity in the fossil query database respectively. They represent the numerical values of the influence degrees of the edge contour similarity, texture similarity, and high-dimensional feature similarity on the correlation scoring coefficient, and can be directly obtained from the fossil query database during use. For example, the edge contour similarity forms a mapping set with the compensation value corresponding to the preset edge contour similarity in the fossil query database, and the compensation value corresponding to the edge contour similarity is obtained by inputting the edge contour similarity into the mapping set; the texture similarity forms a mapping set with the compensation value corresponding to the preset texture similarity in the fossil query database, and the compensation value corresponding to the texture similarity is obtained by inputting the texture similarity into the mapping set; the high-dimensional feature similarity forms a mapping set with the compensation value corresponding to the preset high-dimensional feature similarity in the fossil query database, and the compensation value corresponding to the high-dimensional feature similarity is obtained by inputting the high-dimensional feature similarity into the mapping set, where the mapping relationship is a many-to-one or one-to-one relationship. In this embodiment, the value range of the compensation value is between 0 and 1.

[0057] Finally, sort the correlation scoring coefficients of each bio-standardized fossil image from large to small to obtain the maximum correlation scoring coefficient; obtain the maximum correlation scoring threshold from the preset fossil query database; compare the maximum correlation scoring coefficient with the maximum correlation scoring threshold. If the maximum correlation scoring coefficient is less than the maximum correlation scoring threshold, it is prompted that there are no similar fossils in the bio-standardized fossil database. If the maximum correlation scoring coefficient is greater than or equal to the maximum correlation scoring threshold, the bio-standardized fossil corresponding to the maximum correlation scoring coefficient is used as the fossil type to be identified.

[0058] In this embodiment, through multi-dimensional feature extraction, the key information of fossils can be comprehensively and carefully extracted, effectively reducing the influence caused by image cracks or other interference factors, thereby improving the accuracy and reliability of fossil identification; by comparing with the maximum correlation scoring threshold in the preset fossil query database, the judgment standard of similarity can be dynamically adjusted in each identification. When the maximum similarity of the fossil to be identified is higher than the threshold, the method automatically prompts the type of similar fossils; when it is lower than the threshold, the method will prompt that no matching fossil species are found in the database. Such a feedback mechanism ensures the flexibility and accuracy of the evaluation process.

[0059] Such as Figure 3As shown in the figure, it is a schematic structural diagram of an online query system for a biological standardized fossil database provided by an embodiment of the present application. The online query system for a biological standardized fossil database provided by an embodiment of the present application includes: a lighting impact reduction processing module, a crack impact reduction processing module, an online query module, and a fossil query database; among them, the lighting impact reduction processing module is used to divide the fossil image to be identified into a crack area and a non-crack area, and perform lighting impact reduction processing on the fossil image to be identified according to the light condition data of the fossil image to be identified and the glossiness of the fossil to be identified. The lighting impact reduction processing is used to reduce the impact of lighting differences on the accuracy of feature extraction of the fossil image to be identified; the crack impact reduction processing module is used to perform crack impact reduction processing on the crack area of the fossil image to be identified after the lighting impact reduction processing according to the crack intensity parameter of the fossil image to be identified after the lighting impact reduction processing. The crack impact reduction processing is used to reduce the impact of crack differences on the similarity evaluation of the fossil image to be identified; the online query module is used to evaluate the similarity between the fossil image to be identified after the crack impact reduction processing and each biological standardized fossil image in the biological standardized fossil database, and divide the types of fossils to be identified according to the similarity evaluation result.

[0060] In summary, in the embodiment of the present application, the lighting impact reduction processing is performed on the image through the light condition data and the fossil glossiness, reducing the interference of lighting differences on the feature extraction of the fossil image; by performing crack impact reduction processing on the crack area, reducing the interference of crack differences on the similarity evaluation; by evaluating the similarity between the image after the crack impact reduction processing and each standardized fossil image in the biological standardized fossil database, the similarity between the fossil to be identified and the fossil image in the database can be accurately compared, and according to the similarity evaluation result, the types of fossils to be identified can be effectively divided, achieving the improvement of the accuracy and automation level of fossil identification.

[0061] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0062] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and combinations of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device generate means for implementing the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0063] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including instruction means that implement the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0064] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operational steps are performed on the computer or other programmable device to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0065] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0066] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these changes and modifications.

Claims

1. An online query method for a biological standardized fossil database, characterized in that, Including the following steps: Dividing the fossil image to be identified into a crack area and a non-crack area, and performing light influence reduction processing on the fossil image to be identified according to the light condition data and the glossiness of the fossil image to be identified. The light influence reduction processing is used to reduce the influence of light difference on the accuracy of feature extraction of the fossil image to be identified; Performing crack influence reduction processing on the crack area of the fossil image to be identified after the light influence reduction processing according to the crack intensity parameter of the fossil image to be identified after the light influence reduction processing. The crack influence reduction processing is used to reduce the influence of crack difference on the similarity evaluation of the fossil image to be identified; Performing similarity evaluation on the fossil image to be identified after the crack influence reduction processing and each biostandard fossil image in the biostandard fossil database, and dividing the types of fossils to be identified according to the similarity evaluation results.

2. The online query method for a biological standardized fossil database according to claim 1, wherein: The step of performing light influence reduction processing on the fossil image to be identified according to the light condition data and the glossiness of the fossil image to be identified includes: Matching the glossiness of the fossil image to be identified with the light disturbance evaluation threshold corresponding to each glossiness value in the preset fossil query database to obtain the light disturbance evaluation threshold corresponding to the glossiness of the fossil image to be identified; Quantifying the influence of light on the quality of the fossil image to be identified according to the light condition data of the fossil image to be identified to obtain a light disturbance evaluation index; Comparing the light disturbance evaluation index with the light disturbance evaluation threshold. If the light disturbance evaluation index is less than or equal to the light disturbance evaluation threshold, no additional operation is performed. Otherwise, the fossil image to be identified is subjected to light influence reduction processing according to the deviation light disturbance evaluation index, and the deviation light disturbance evaluation index represents the deviation situation between the light disturbance evaluation index and the light disturbance evaluation threshold.

3. The online query method for a biological standardized fossil database according to claim 2, characterized in that: The light condition data includes the high-light pixel ratio, the standard gray difference, and the shadow coverage rate; The step of quantifying the influence of light on the quality of the fossil image to be identified according to the light condition data of the fossil image to be identified to obtain a light disturbance evaluation index includes: Obtaining light condition data reference data from the preset fossil query database, specifically including: the critical high-light pixel ratio, the critical standard gray difference, and the critical shadow coverage rate; Performing ratio approximation operations on the high-light pixel ratio, the standard gray difference, and the shadow coverage rate with the critical high-light pixel ratio, the critical standard gray difference, and the critical shadow coverage rate respectively, and then performing weighted coupling processing on the ratio approximation operation results through the light condition data compensation values respectively to obtain a light disturbance evaluation index. The light condition data compensation values include a high-light pixel influence factor, a standard gray difference influence factor, and a shadow coverage rate influence factor. The light disturbance evaluation index represents the quantification data of the combined influence degree of the high-light pixel ratio, the standard gray difference, and the shadow coverage rate on the quality of the fossil image to be identified.

4. The online query method for a biological standardized fossil database according to claim 2, wherein: The step of performing light influence reduction processing on the fossil image to be identified according to the deviation light disturbance evaluation index includes: Obtaining a first deviation light disturbance evaluation threshold and a second deviation light disturbance evaluation threshold from the preset fossil query database; Compare the deviation light disturbance evaluation index with the first deviation light disturbance evaluation threshold and the second deviation light disturbance evaluation threshold respectively. If the deviation light disturbance evaluation index is less than the first deviation light disturbance evaluation threshold, turn on the restricted histogram equalization, and at the same time enhance the contrast of the crack area according to the deviation light disturbance evaluation index. If the deviation light disturbance evaluation index is greater than or equal to the first deviation light disturbance evaluation threshold and less than the second deviation light disturbance evaluation threshold, decompose the fossil image to be identified into high-frequency components and low-frequency components, apply homomorphic filtering to the low-frequency components, and perform adaptive threshold denoising on the high-frequency components. If the deviation light disturbance evaluation index is greater than or equal to the second deviation light disturbance evaluation threshold, separate the illumination and reflection components, eliminate the reflection component from the fossil image to be identified, and at the same time turn on the adaptive homomorphic filtering.

5. The online query method for a biological standardized fossil database according to claim 4, characterized in that: The step of enhancing the contrast of the crack area according to the deviation light disturbance evaluation index includes: Obtain the contrast enhancement value corresponding to each deviation light influence area from the preset fossil query database; Match the deviation light disturbance evaluation index with each deviation light influence area to obtain the deviation light influence areas corresponding to the deviation light disturbance evaluation index, so as to obtain the contrast enhancement value corresponding to the deviation light disturbance evaluation index; Process the contrast of the fossil image to be identified according to the contrast enhancement value to reduce the influence of illumination on the quality of the crack area of the fossil image to be identified.

6. The online query method of a biological standardized fossil database according to claim 1, characterized in that: The step of reducing the influence of cracks on the crack area of the fossil image to be identified after reducing the influence of illumination according to the crack strength parameter after reducing the influence of illumination includes: Quantify the influence of cracks on the fossil image to be identified after reducing the influence of illumination according to the crack strength parameter to obtain the crack strength index; Obtain the first crack strength index threshold and the second crack strength index threshold from the preset fossil query database; Compare the crack strength index with the first crack strength index threshold and the second crack strength index threshold respectively. If the crack strength index is less than the first crack strength index threshold, fill and repair the crack area of the fossil image to be identified after reducing the influence of illumination according to the adjacent pixels of each crack area. If the crack strength index is greater than or equal to the first crack strength index threshold and less than the second crack strength index threshold, extract the texture features and pixel information from around the crack area to fill and repair the crack area of the fossil image to be identified after reducing the influence of illumination. If the crack strength index is greater than or equal to the second crack strength index threshold, prompt that the cracks in the fossil to be identified cannot be recognized.

7. The online query method for a biological standardized fossil database according to claim 6, characterized in that: The crack strength parameter includes the crack area ratio, the crack depth ratio, and the number of cracks; The step of quantifying the influence of cracks on the fossil image to be identified after reducing the influence of illumination according to the crack strength parameter to obtain the crack strength index includes: Obtain the crack strength parameter reference data from the preset fossil query database, specifically including: the critical crack area ratio, the critical crack depth ratio, and the critical number of cracks; Perform ratio approximation operations on the crack area ratio, crack depth ratio, and number of cracks respectively with the critical crack area ratio, critical crack depth ratio, and critical number of cracks. Then, perform weighted coupling processing on the results of the ratio approximation operations through the crack strength parameter compensation value to obtain the crack strength index. The crack strength parameter compensation value includes the crack area ratio influence factor, crack depth ratio influence factor, and number of cracks influence factor. The crack strength index represents the quantitative data of the influence degree of the crack area ratio, crack depth ratio, and number of cracks on the crack area of the fossil image to be identified after the illumination influence reduction processing.

8. The online query method for a biological standardized fossil database according to claim 1, wherein: The steps of evaluating the similarity between the fossil image to be identified after the crack influence reduction processing and each biological standard fossil image in the biological standard fossil database, and classifying the types of fossils to be identified according to the similarity evaluation results include: Extract the characteristics of the fossil image to be identified after the crack influence reduction processing and the characteristics of each biological standard fossil image. The image characteristics include the fossil edge contour point set, fossil surface texture, and high-dimensional feature vector. Obtain the edge contour correlation scoring coefficient according to the fossil edge contour point set. Obtain the texture correlation scoring coefficient according to the fossil surface texture. Obtain the high-dimensional feature correlation scoring coefficient according to the high-dimensional feature vector. Quantify the similarity between the fossil image to be identified after the crack influence reduction processing and each biological standard fossil image in the biological standard fossil database according to the edge contour correlation scoring coefficient, texture correlation scoring coefficient, and high-dimensional feature correlation scoring coefficient, and obtain the correlation scoring coefficient of each biological standard fossil image. Sort the correlation scoring coefficients of each biological standard fossil image from largest to smallest to obtain the maximum correlation scoring coefficient. Obtain the maximum correlation scoring threshold from the preset fossil query database. Compare the maximum correlation scoring coefficient with the maximum correlation scoring threshold. If the maximum correlation scoring coefficient is less than the maximum correlation scoring threshold, it is prompted that there are no similar fossils in the biological standard fossil database. If the maximum correlation scoring coefficient is greater than or equal to the maximum correlation scoring threshold, the biological standard fossil corresponding to the maximum correlation scoring coefficient is used as the type of fossil to be identified.

9. The online query method for a biological standardized fossil database according to claim 8, wherein: The steps of quantifying the similarity between the fossil image to be identified after the crack influence reduction processing and each biological standard fossil image in the biological standard fossil database according to the edge contour correlation scoring coefficient, texture correlation scoring coefficient, and high-dimensional feature correlation scoring coefficient, and obtaining the correlation scoring coefficient of each biological standard fossil image include: Use the similarity compensation value to perform weighted coupling processing on the edge contour correlation scoring coefficient, texture correlation scoring coefficient, and high-dimensional feature correlation scoring coefficient of each biological standard fossil image respectively to obtain the correlation scoring coefficient of each biological standard fossil image. The similarity compensation value includes the edge contour similarity influence factor, texture similarity influence factor, and high-dimensional feature similarity influence factor. The correlation scoring coefficient of each biological standard fossil image represents the quantitative data of the influence degree of the edge contour correlation scoring coefficient, texture correlation scoring coefficient, and high-dimensional feature correlation scoring coefficient on the similarity degree between the fossil to be identified and each biological standard fossil.

10. An online query system for a biological standardized fossil database, characterized in that, It includes a light influence reduction processing module, a crack influence reduction processing module, an online query module, and a fossil query database; Among them, the light influence reduction processing module is used to divide the fossil image to be identified into a crack area and a non-crack area, and perform light influence reduction processing on the fossil image to be identified according to the light condition data of the fossil image to be identified and the glossiness of the fossil to be identified. The light influence reduction processing is used to reduce the influence of light differences on the accuracy of feature extraction of the fossil image to be identified; The crack influence reduction processing module is used to perform crack influence reduction processing on the crack area of the fossil image to be identified after the light influence reduction processing according to the crack intensity parameter of the fossil image to be identified after the light influence reduction processing. The crack influence reduction processing is used to reduce the influence of crack differences on the similarity evaluation of the fossil image to be identified; The online query module is used to evaluate the similarity between the fossil image to be identified after the crack influence reduction processing and each bio-standard fossil image in the bio-standard fossil database, and divide the types of fossils to be identified according to the similarity evaluation results.

Images