Plant vein custom identification printing method and system based on image identification

Abstract

The invention relates to the field of plant vein printing. Fabric and paper can be made into fine color textures by applying a traditional plant dyeing method, but a traditional dyeing process is difficult to meet industrial production requirements, and industrial pattern printing and dyeing lack of natural textures and low pattern change freedom degree are one of main factors restricting artistry and individuation. The invention provides a plant vein user-defined identification printing method and system based on image identification, and aims to improve the production efficiency, maintain the natural aesthetic feeling and fine texture of veins, extract vein images from multiple positions, obtain more comprehensive vein information, and ensure that the vein images of the whole leaf can be synthesized into the vein images of the whole leaf. And any important vein features are not omitted. By acquiring the high-quality microscopic image in real time, the high definition and details of the vein can be kept, so that the real vein texture can be re-carved during printing. A user can customize the vein image according to own requirements, and batch recarving can be realized through an automatic processing flow.

Description

Technical Field

[0001] The present application relates to the technical field of plant vein printing, and in particular to a method and system for customizing plant vein identification and printing based on image recognition. Background Art

[0002] Traditional hand-dying is an ancient art form that creates unique patterns and colors by applying plant leaves, flowers or other natural materials directly to fabrics. Traditional hand-dying not only demonstrates the creativity of artists, but also inherits rich culture and history. However, it is relatively inefficient in the face of modern industrial needs because it is time-consuming and labor-intensive and difficult to mass-produce.

[0003] Plant leaf veins are a complex network structure inside the leaves that is responsible for transporting water and nutrients from the roots of the plant to the leaves, and is also responsible for photosynthesis on the leaves. Leaf veins are usually composed of xylem and phloem. The xylem is responsible for the transport of water and minerals, while the phloem is responsible for the transport of organic matter. The morphology and structure of leaf veins are unique in different plant species, which makes leaf veins an important feature in plant taxonomy. Characteristics such as the shape, size, and branching pattern of leaf veins can be used to distinguish different plant species. In addition, the development and structure of leaf veins are also affected by environmental factors such as light, water, temperature, etc., so the study of leaf veins is of great significance for understanding the growth and adaptation of plants to the environment.

[0004] Existing technologies face a series of challenges when dealing with plant vein reproduction. Since the veins of each leaf are unique and have certain artistic and educational significance, existing digital printing technologies can print clear images, but it is often difficult to achieve ideal accuracy for structures such as veins with complex textures and details, especially for fine branches and reticular structures. It is difficult to capture fine branches and reticular structures, and it is easy to ignore some tiny or broken veins or include parts that do not belong to the veins, resulting in incomplete reproduction images or misjudgment of features.

[0005] Therefore, the prior art has defects and needs to be improved. Summary of the invention

[0006] In order to solve one or several problems in the prior art, the main purpose of the present application is to provide a method and system for customizing plant vein identification and printing based on image recognition.

[0007] In order to achieve the above-mentioned invention object, the present application proposes a plant vein custom identification printing method based on image recognition, the method comprising:

[0008] Acquire a microscopic image of the leaf in real time, and determine whether the microscopic image meets the output conditions;

[0009] When the microscopic image meets the output condition, extracting a leaf vein image set in the microscopic image from multiple positions according to a preset acquisition standard;

[0010] Receive printing instructions from the user in real time;

[0011] When receiving a printing instruction from the user terminal, parsing the printing instruction to determine the printing requirements of the leaf vein image;

[0012] According to the printing requirement, selecting a leaf vein image that meets the printing requirement from a set of leaf vein images at multiple locations;

[0013] According to the screening result, each screened leaf vein image is fused, and the fused leaf vein image is converted into a vector image, wherein the leaf vein image fusion is used to synthesize the leaf vein image of the whole leaf;

[0014] The vector image is used to generate a printing pattern file.

[0015] The embodiment of the present application also provides a plant vein custom recognition and printing system based on image recognition, comprising:

[0016] An acquisition module, used for acquiring a microscopic image of the leaf in real time and determining whether the microscopic image meets the output conditions;

[0017] An extraction module, used for extracting a leaf vein image set in the microscopic image from multiple positions according to a preset acquisition standard when the microscopic image meets the output condition;

[0018] A receiving module, used for receiving a printing instruction from a user terminal in real time;

[0019] A parsing module, for parsing the printing instruction and determining the printing requirements of the leaf vein image when receiving the printing instruction from the user terminal;

[0020] A screening module, used for screening leaf vein images that meet the printing requirements from leaf vein images at multiple locations according to the printing requirements;

[0021] A fusion module, used for fusing each filtered leaf vein image according to the filtering result, and converting the fused leaf vein image into a vector image, wherein the leaf vein image fusion is used to synthesize the leaf vein image of the whole leaf;

[0022] A generating module is used to generate a printing pattern file from the vector image.

[0023] The present application also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0024] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0025] The plant vein custom recognition printing method and system based on image recognition of the embodiment of the present application can obtain more comprehensive vein information by extracting vein images from multiple positions, ensuring that no important vein features are missed when synthesizing the vein image of the entire leaf. By acquiring high-quality microscopic images in real time, the high definition and details of the veins can be retained, so that the real vein texture can be reproduced when printing. The function of receiving the printing instructions of the user end in real time is provided, so that the user can customize the vein image according to his own needs, increasing the flexibility and user experience of the printing process. By screening and fusing the vein images of multiple positions, a complete and coherent vein image can be generated, which improves the accuracy and artistic value of the printed pattern. The fused vein image is converted into a vector image, supporting lossless magnification and editing, ensuring the quality and customizability of the printed pattern. Through the automated processing flow, batch reproduction can be achieved, which improves the printing efficiency, while ensuring the consistency and accuracy of each printed pattern, and realizing batch printing of veins of the same leaf. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A schematic diagram of a process of a method for customizing plant vein identification and printing based on image recognition according to an embodiment of the present application;

[0027] Figure 2 A schematic diagram of a process of a method for customizing plant vein identification and printing based on image recognition according to an embodiment of the present application;

[0028] Figure 3 This is a schematic structural block diagram of a plant vein custom recognition and printing system based on image recognition according to an embodiment of the present application;

[0029] Figure 4 A schematic block diagram of the structure of a computer device according to an embodiment of the present application.

[0030] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0032] Reference Figure 1 In an embodiment of the present application, a method for customizing plant vein identification and printing based on image recognition is provided, and the method comprises:

[0033] S1. Acquire a microscopic image of a leaf in real time, and determine whether the microscopic image meets output conditions;

[0034] S2. When the microscopic image meets the output condition, extracting a leaf vein image set in the microscopic image from multiple positions according to a preset acquisition standard;

[0035] S3, receiving the printing instruction from the user in real time;

[0036] S4, when receiving a printing instruction from the user terminal, parsing the printing instruction to determine the printing requirement of the leaf vein image;

[0037] S5. According to the printing requirement, selecting a leaf vein image that meets the printing requirement from the leaf vein image set at multiple positions;

[0038] S6. According to the screening result, each screened leaf vein image is fused, and the fused leaf vein image is converted into a vector image, wherein the leaf vein image fusion is used to synthesize the leaf vein image of the whole leaf;

[0039] S7, generating a printing pattern file from the vector image.

[0040] As described in the above steps S1-S2, a microscopic image of the leaf is acquired in real time by a microscope, and then the image quality, such as clarity, contrast, etc., is evaluated using image processing technology. This step ensures that only high-quality microscopic images can be used for subsequent processing, thereby improving the quality of the final printed pattern. In the microscopic images that meet the output conditions, vein images are extracted from multiple positions according to preset acquisition standards. These standards include the clarity, integrity and diversity of the veins. By extracting vein images from multiple positions, more comprehensive vein information can be ensured, thereby improving the diversity and accuracy of the printed pattern. Under a microscope, due to the limitation of the field of view, one shot may not cover the characteristics of all veins on the leaf. Therefore, by extracting vein images from multiple positions, more comprehensive vein information can be ensured. The purpose of this is to ensure that the vein image of the entire leaf can be synthesized in the final printed pattern, rather than relying solely on the image of a single position. The vein network of a leaf may be very complex, and shooting from different angles and positions can capture different branches and details of the veins. By fusing these vein images from different locations together, a more complete and accurate vein image can be generated, which helps to present the complete structure and details of the veins when printing.

[0041] As described in the above steps S3-S5, the user can send printing instructions in real time through the interface, such as selecting a specific leaf vein image, adjusting printing parameters, etc. Receiving printing instructions in real time allows users to flexibly customize printing requirements, thereby improving the flexibility of the printing process and user experience. Parse the received printing instructions to understand the user's needs, such as the selection of leaf vein images, print size, color, etc. By parsing the printing instructions, it can be ensured that subsequent processing can meet the user's needs and improve the accuracy of printing. According to the user's printing needs, leaf vein images that meet the requirements are screened out from the leaf vein images in multiple locations. Through screening, it can be ensured that the final printed leaf vein image meets the specific needs of the user and improves the pertinence of printing.

[0042] As described in the above steps S6-S7, the filtered vein images are fused to synthesize the vein image of the entire leaf, and then the fused image is converted into a vector format. Through fusion and vectorization, the continuity and integrity of the vein image can be ensured, while supporting lossless enlargement and editing, thereby improving the quality of the printed pattern. The vector image is converted into a format suitable for printing, such as PDF, EPS, etc., for use by the printer. The generation of a print pattern file enables the vein image to be accurately printed by the printer, realizing batch printing of veins of the same leaf, and also ensuring the quality and efficiency of printing.

[0043] As described above, by extracting vein images from multiple positions, more comprehensive vein information can be obtained to ensure that no important vein features are missed when synthesizing the vein image of the entire leaf. By acquiring high-quality microscopic images in real time, the high definition and details of the veins can be retained, so that the real vein texture can be reproduced when printing. Providing the function of receiving printing instructions from the user end in real time allows users to customize vein images according to their needs, increasing the flexibility of the printing process and user experience. By screening and fusing vein images at multiple positions, a complete and coherent vein image can be generated, which improves the accuracy and artistic value of the printed pattern. The fused vein image is converted into a vector image, supporting lossless magnification and editing, ensuring the quality and customizability of the printed pattern. Through the automated processing flow, batch reproduction can be achieved, which improves printing efficiency, while ensuring the consistency and accuracy of each printed pattern, and realizing batch printing of veins of the same leaf.

[0044] In one embodiment, the fabric and paper can produce delicate color texture by using traditional plant dyeing methods, but the traditional dyeing process is inefficient and difficult to meet the needs of industrial production, and the industrial pattern printing and dyeing lacks natural texture, and the low degree of freedom of pattern change is also one of the main factors restricting its artistic and personalized. In the modern textile and paper production process, the traditional plant dyeing method is favored by its natural and environmentally friendly characteristics. However, its production efficiency is relatively low and it is difficult to meet the needs of large-scale production. This embodiment proposes an innovative solution that combines traditional plant dyeing with modern digital printing and dyeing technology to improve dyeing efficiency and artistic expression. In order to further meet the market's demand for personalization, this embodiment can also introduce a customizable design platform, and consumers can choose colors and patterns according to their personal preferences, and even participate in the design. To this end, the method of the present invention proposes a plant leaf vein custom recognition printing method and system based on image recognition, which aims to improve production efficiency while retaining the natural beauty and delicate texture of the leaf veins. By extracting leaf vein images from multiple positions, more comprehensive leaf vein information can be obtained to ensure that when synthesizing the leaf vein image of the entire leaf, no important leaf vein features are omitted. By acquiring high-quality microscopic images in real time, the high definition and details of the leaf veins can be retained, so that the real leaf vein texture can be reproduced when printing. The function of receiving printing instructions from the user end in real time is provided, so that users can customize leaf vein images according to their needs, and batch reproduction can be achieved through automated processing.

[0045] Reference Figure 2 In one embodiment, before the step of fusing each filtered leaf vein image, the method comprises:

[0046] S51, identifying each leaf vein image after screening, and determining whether there is an image defect in the leaf vein image;

[0047] S52, when an image defect appears in the leaf vein image, marking the defect position in the leaf vein image, and analyzing the type of the defect appearing in the leaf vein image;

[0048] S53, when the defect type is that adjacent leaf veins are too close to each other, resulting in blurred image;

[0049] S54, inputting the defect type and the leaf vein image into a preset segmentation model respectively, segmenting adjacent leaf veins by the segmentation model, and identifying target features of each leaf vein, wherein the target features include branch points, texture, color, and direction of the leaf vein;

[0050] S55, filling the boundaries of the segmented leaf veins according to the target features through the segmentation model, and outputting the processed leaf vein image;

[0051] S56, performing fusion based on the processed leaf vein image.

[0052] As described in the above steps, each leaf vein image is analyzed by image processing technology to find possible defects, such as blur, breakage, noise, etc. Since the leaf vein image may be affected by factors such as shooting conditions and microscope imaging quality, identifying defects is to ensure the quality of the fused image. Timely discovery and marking of defects provides a basis for subsequent repair and improves the quality of the final fused image. The defects found are marked and analyzed according to their appearance characteristics to determine the type of defects. Accurate defect position marking and type analysis help to select the appropriate repair method, thereby improving the repair effect. Through accurate marking and analysis, corresponding repair measures can be taken for different types of defects, ensuring the accuracy of repair and the integrity of the image. Identify the image blur problem caused by adjacent leaf veins being too close, and take repair measures for this specific type of defect. The adjacent leaf veins in the leaf vein image are too close. It is a common defect type that has a direct impact on the printing quality, so it needs to be specially processed. By specially processing this type of defect, the clarity of the leaf vein image can be restored and the visual effect of the printed pattern can be improved. Use the pre-trained segmentation model to segment the adjacent leaf veins and extract the target features of each leaf vein. The segmentation model can accurately identify the boundaries of leaf veins and extract key features, which is helpful for subsequent repair and fusion. Through segmentation and feature extraction, the blurred areas caused by the close proximity of adjacent leaf veins can be accurately repaired and the details of the leaf vein image can be restored. According to the segmented leaf vein boundaries and target features, the boundaries are filled to repair the blurred areas. The filling process can fill the blank areas after segmentation and restore the integrity and continuity of the leaf vein image. Through the filling process, the blurred areas caused by the close proximity of adjacent leaf veins can be repaired to improve the clarity and print quality of the image. The repaired leaf vein image is synthesized into a whole using fusion technology to obtain a complete leaf vein image. The repaired leaf vein image may have discontinuous boundaries or overlapping areas. Fusion can solve these problems and generate high-quality printed patterns. Through fusion processing, a complete and coherent leaf vein image can be synthesized to improve the integrity and artistic value of the printed pattern.

[0053] In one embodiment, after the steps of marking the defect position in the leaf vein image when an image defect appears in the leaf vein image, and analyzing the type of the defect appearing in the leaf vein image, the method further comprises:

[0054] When the defect type is shadow and / or reflection on the leaf vein, identifying the illumination information of the defect position and the reflection characteristics of the leaf vein surface;

[0055] Determining light radiation parameters on the surface of the leaf veins according to the illumination information and the reflection characteristics;

[0056] Inputting the light radiation parameters and the light information into a preset light correction model, and generating light adjustment parameters through the light correction model;

[0057] Segmenting the area where the leaf veins appear shadows and / or reflections, adjusting the illumination information based on illumination adjustment parameters, and re-correcting the leaf vein image at the defective position using the adjusted illumination information;

[0058] Fusion is performed based on the corrected leaf vein images.

[0059] As described above, by analyzing the shadow and reflective areas in the image, the illumination information and reflective features of the vein surface are extracted. Shadows and reflections will interfere with the visual effect of the vein image and affect subsequent recognition and fusion. Accurately identifying the illumination information of the defective position is helpful for subsequent illumination correction and improves the visual effect of the image. According to the extracted illumination information and reflective features, the light radiation parameters of the vein surface, such as illumination intensity and direction, are calculated. The light radiation parameters are the input of the illumination correction model and are crucial to the correction effect. Determining accurate light radiation parameters can improve the accuracy of illumination correction and reduce the impact of shadows and reflections on the image. Using the pre-trained illumination correction model, the light radiation parameters and illumination information are used as input to generate illumination adjustment parameters. The illumination correction model can automatically adjust the illumination parameters according to the input information to achieve image correction. Generating illumination adjustment parameters can automatically correct shadows and reflections in the image and improve the visual effect of the image. The shadow and reflective areas are segmented, and then the illumination information is adjusted according to the illumination adjustment parameters to achieve re-correction of the vein image at the defective position. Segmentation and adjustment of illumination information can correct defective areas in a targeted manner and improve the correction effect. By re-correcting the vein image at the defective position, shadows and reflections can be removed to improve the visual effect of the image. Fusion technology is used to synthesize the corrected vein images into a whole to obtain a complete vein image. The corrected vein image may have discontinuous boundaries or overlapping areas, and fusion can solve these problems and generate high-quality printed patterns. Through fusion processing, a complete and coherent vein image can be synthesized to improve the integrity and artistic value of the printed pattern.

[0060] In one embodiment, the method of fusing each filtered leaf vein image comprises:

[0061] Obtain each filtered leaf vein image, and adapt the position of each leaf vein image;

[0062] According to the adaptation results, mark the overlapping areas between different images;

[0063] Identify the target features of the leaf veins in the overlapping area, construct an image coordinate system based on the target features, input the target features into the image coordinate system, and obtain a fusion position of the same overlapping area, wherein the target features include a branch point, texture, and direction of the leaf veins;

[0064] The overlapping areas are merged according to the fusion positions, and the edges of the merged leaf veins are corroded and expanded;

[0065] Based on the processing results, the fusion of leaf vein images is completed.

[0066] As mentioned above, image processing technology is used to ensure that each leaf vein image is spatially aligned with other images. Leaf vein images at different locations need to be accurately aligned so that they can cover the entire leaf when fused. The adapted images can be fused together more accurately to form a complete leaf vein image. Identify and mark the overlapping areas between different images, which are the key part of fusion. The overlapping areas contain continuous information of leaf veins and need to be accurately fused together. Marking the overlapping areas helps to fuse the images more accurately and ensure the continuity of leaf veins. Identify the target features of leaf veins in the overlapping areas, such as branch points, textures, and directions, and construct an image coordinate system based on these features. The target features are the basis for leaf vein image fusion, and constructing an image coordinate system helps to maintain the structure and continuity of leaf veins during the fusion process. The image coordinate system constructed based on the target features helps to fuse the overlapping areas more accurately to form a complete leaf vein image. The overlapping areas are fused according to the fusion position, and then the edges of the fused leaf veins are eroded and expanded to enhance the continuity and clarity of the image. The erosion and expansion processing can fill the gaps in the fused area and enhance the edge details of the leaf veins. Through corrosion and expansion processing, a clearer and more coherent leaf vein image can be generated to improve the quality of the printed pattern. The processed leaf vein images are synthesized into a whole using fusion technology to obtain a complete leaf vein image. Fusion processing can combine multiple leaf vein images into a continuous leaf vein image to form a complete leaf vein pattern. Through fusion processing, a complete and coherent leaf vein image can be generated.

[0067] In a specific embodiment, the leaf can be photographed at 100 times using an optical microscope to ensure that the image clarity is at least 200 dpi. A high-magnification microscope can capture the subtle features of the leaf veins, and a high-definition image is helpful for subsequent processing and printing. A screening standard is set, such as the leaf vein clarity is not less than 70%, and the image has no obvious distortion. The screening standard ensures that only high-quality images can be used for subsequent processing, thereby improving the quality of the final printed pattern. An image defect detection algorithm based on deep learning is used to detect defects such as blur, breakage, and noise in the image. For blur caused by adjacent leaf veins being too close, the similarity threshold of the segmentation model is set to 0.9 to ensure the accuracy of the leaf vein features. Through defect detection and repair, the image quality can be improved to ensure the clarity and accuracy of the printed pattern. An image fusion algorithm based on multi-scale fusion is used, and the fusion weight is set to 0.8 to balance the clarity and details of the image. The fusion process can synthesize a complete and coherent leaf vein image, thereby improving the integrity and artistic value of the printed pattern. The fused leaf vein image is converted into a vector format, and the conversion accuracy is set to 300 dpi to ensure the printing quality. Vectorization supports lossless enlargement and editing, ensuring the quality and customizability of the printed pattern. Generate a print pattern file from the vector image, set the print size to A4, and the color mode to CMYK. Generate a print pattern file so that the leaf vein image can be accurately printed by the printer.

[0068] In one embodiment, after the step of extracting the leaf vein image set in the microscopic image from multiple positions, the method further comprises:

[0069] Calculating a differential image between the leaf vein image and an original microscopic image corresponding to the leaf vein image;

[0070] According to the calculation result, obtaining the missing features in the difference image;

[0071] Identifying whether the missing features in the differential image meet the condition of leaf vein features;

[0072] Extracting the shape, texture and size information of the missing feature, inputting the shape, texture and size information of the missing feature and the leaf vein image into a preset matching model respectively, and analyzing the similarity coefficient between the missing feature and the leaf vein feature through the matching model;

[0073] When the similarity coefficient is greater than a preset similarity threshold, it is determined that the missing feature meets the condition of the leaf vein feature;

[0074] Based on the determination result, the missing features are filled into the leaf vein image to obtain an adjusted leaf vein image.

[0075] As described above, by calculating the difference image between the vein image and its corresponding original microscopic image, the difference between the two, that is, the possible missing features, can be found. The difference image can intuitively show the difference between the vein image and the original image, which helps to find the missing features. The difference image provides a basis for identifying the missing features and provides a foundation for subsequent repair and fusion. By analyzing the difference image, the features that may be missed, such as tiny vein branches or broken parts, can be extracted. The missing features may be an important component of the vein image and need to be identified and repaired. Obtaining the missing features helps to improve the integrity and accuracy of the vein image. By analyzing the shape, texture, size and other information of the missing features, it is determined whether it meets the conditions of the vein features. Only the missing features that meet the vein feature conditions need to be repaired and fused. Identifying the missing features that meet the conditions helps to improve the integrity and accuracy of the vein image. Extract the key information of the missing features and input them into the matching model together with the vein image, and analyze the similarity coefficient between the two through the model. The matching model can automatically determine the similarity between the missing features and the vein features, which helps to accurately identify and repair the missing features. By analyzing the similarity coefficient through the matching model, it is possible to accurately determine whether the missing features need to be repaired and fused. Set a similarity threshold. When the similarity coefficient is greater than the threshold, the missing features are determined to meet the conditions of the vein features. The similarity threshold can be used as a judgment standard to ensure that the similarity between the missing features and the vein features reaches a certain level. Judging whether the missing features meet the conditions of the vein features based on the similarity coefficient helps to improve the integrity and accuracy of the vein image. Based on the judgment result of the similarity coefficient, the missing features that meet the conditions are filled into the vein image. Filling the missing features can repair the defects in the vein image and improve the integrity and accuracy of the image. By filling the missing features, a more complete and accurate vein image can be obtained, providing a high-quality foundation for subsequent fusion and printing.

[0076] In one embodiment, after the step of filling the missing features into the leaf vein image, the method further comprises:

[0077] Determining whether the missing feature satisfies the condition of the broken feature of the leaf vein;

[0078] Extracting edge features of the missing features, based on whether there are irregular fracture traces on the edges of the missing features;

[0079] When there are irregular fracture marks on the edge of the missing feature, analyzing whether the texture feature of the missing feature is continuous with the target leaf vein;

[0080] When the texture feature of the omitted feature is continuous with the target leaf vein, obtaining whether the cross-sectional width of the omitted feature matches the cross-sectional width of the target leaf vein;

[0081] When the cross-sectional width of the omitted feature matches the cross-sectional width of the target leaf vein, analyzing whether the direction of the omitted feature is consistent with the direction of the target leaf vein;

[0082] When the direction of the missing feature is consistent with the direction of the target leaf vein, it is determined that the missing feature meets the condition of the broken feature of the leaf vein;

[0083] Based on the fact that the missing features satisfy the broken features of leaf veins, the missing features are filled onto the target leaf veins to obtain an adjusted leaf vein image.

[0084] As described above, by analyzing the shape, texture, size and other information of the missing feature, it is determined whether it meets the characteristics of leaf vein breakage, such as irregular edges, texture breakage, etc. Leaf vein breakage features may be easily overlooked in the image processing process. Accurately judging whether the missing feature meets the conditions of leaf vein breakage features is helpful to improve the integrity and accuracy of the leaf vein image. By extracting the edge features of the missing feature, it can be determined whether it has irregular breakage marks, such as edge irregularity, breakage, etc. Irregular breakage marks are typical manifestations of leaf vein breakage features, and extracting edge features helps to accurately determine whether the missing feature meets the conditions of leaf vein breakage features. By analyzing the texture features of the missing feature, it is determined whether it has continuity with the target leaf vein, such as texture continuity and similarity. Texture features are important manifestations of leaf vein breakage features. Analyzing texture features helps to accurately determine whether the missing feature meets the conditions of leaf vein breakage features. By obtaining the cross-sectional width of the missing feature and the cross-sectional width of the target leaf vein, it is determined whether the two match. Determining whether the cross-sectional widths match helps to accurately determine whether the missing feature meets the conditions of leaf vein breakage features. By analyzing whether the direction of the missing feature is consistent with the direction of the target leaf vein, it is determined whether the two match. Based on the above analysis, it is determined whether the missing feature meets the conditions of the leaf vein break feature. Accurately judging whether the missing feature meets the conditions of the leaf vein break feature helps to improve the integrity and accuracy of the leaf vein image. According to the judgment result, the missing features that meet the conditions are filled onto the target leaf vein to obtain the adjusted leaf vein image.

[0085] It should be added that from the perspective of art and leaf biology, leaf printing is not just a superficial reproduction process, but an opportunity for in-depth research and expression. Each feature in the leaf, such as veins, stomata and surface texture, plays an important role in the growth and adaptation of plants. Through precise printing technology, these tiny details can be recorded to help understand the physiological mechanisms and ecological adaptability of plants. By observing the complex structure of leaves, artworks can better convey this beauty. Differential images can help to highlight details that are not captured in leaf vein images. Noise filtering effectively distinguishes important features from background noise, helping subsequent processing to focus on key areas. Using differential images to extract missing features helps to identify those tiny or broken leaf vein parts and ensure the integrity of the replica. Integrating missing features into the replica image enriches the data set and improves accuracy. By setting clear feature conditions, the possibility of misjudgment is reduced and it is ensured that the extracted features are indeed related to leaf veins. Allowing the system to learn and adapt to the unique structure of different leaves enhances the generalization ability of the model. Detailed feature description helps comparison and classification, ensuring that the missing parts form an effective correspondence with known leaf vein features.

[0086] In one embodiment, the method of converting the fused leaf vein image into a vector image includes:

[0087] Identifying branch points of the fused leaf vein image, and dividing the leaf vein image into a plurality of regions according to the branch points;

[0088] Extract the vein features of each segmented area;

[0089] Convert each extracted leaf vein feature into a vector format to obtain a vector image block;

[0090] Performing a first preprocessing on each vector image block, and synthesizing the first preprocessed vector image blocks;

[0091] Based on the synthesis result, the vector image is subjected to a second preprocessing as a whole to obtain a final vector image.

[0092] As described above, the branch points in the vein image are identified by image processing technology, and these branch points are the key to segmenting the vein image. The branch points in the vein image are the key part of the vein structure, and the vein image can be segmented more accurately by identifying these points. Accurate segmentation of the vein image helps the subsequent feature extraction and vectorization processing, ensuring the accuracy and integrity of the vectorized image. The vein features, such as texture, shape, color, etc., are extracted from each segmented area. The vein features are the basis of vectorization processing, and extracting accurate features helps to generate high-quality vector images. Feature extraction provides key information for vectorization and helps to generate vector images that can accurately reflect the vein structure. The extracted vein features are converted into vector format to generate vector image blocks. The vector format can be losslessly enlarged and edited to ensure printing quality, and the image size can be adjusted more flexibly. The vectorization process generates vector image blocks that can be arbitrarily enlarged without distortion, providing high-quality images for subsequent synthesis and printing. Each vector image block is preprocessed, such as smoothing, sharpening, etc., and then the preprocessed image blocks are synthesized into a whole. Preprocessing can improve image quality, while synthesis can combine multiple image blocks into a complete vein image. Through preprocessing and synthesis, a higher quality and more detailed leaf vein image can be generated, providing a high-quality image foundation for the final printing. Based on the synthesis, the entire vector image is further preprocessed, such as adjusting the color, contrast, etc. Further preprocessing can further improve the overall quality of the image, making it more suitable for printing and display. The vector image after the second preprocessing can better meet the printing needs and improve the artistic value and accuracy of the printing effect.

[0093] The plant vein custom recognition printing method based on image recognition of the present application can obtain more comprehensive vein information by extracting vein images from multiple positions, ensuring that no important vein features are missed when synthesizing the vein image of the entire leaf. By acquiring high-quality microscopic images in real time, the high definition and details of the veins can be retained, so that the real vein texture can be reproduced when printing. The function of receiving the printing instructions of the user end in real time is provided, so that the user can customize the vein image according to his own needs, increasing the flexibility and user experience of the printing process. By screening and fusing the vein images of multiple positions, a complete and coherent vein image can be generated, which improves the accuracy and artistic value of the printed pattern. The fused vein image is converted into a vector image, supporting lossless magnification and editing, ensuring the quality and customizability of the printed pattern. Through the automated processing flow, batch reproduction can be achieved, which improves the printing efficiency, while ensuring the consistency and accuracy of each printed pattern, and realizing batch printing of veins of the same leaf.

[0094] Reference Figure 3In an embodiment of the present application, a plant vein custom recognition and printing system based on image recognition is also provided, comprising:

[0095] Acquisition module 1, used to acquire a microscopic image of a leaf in real time and determine whether the microscopic image meets the output conditions;

[0096] Extraction module 2, used for extracting a leaf vein image set in the microscopic image from multiple positions according to a preset acquisition standard when the microscopic image meets the output condition;

[0097] Receiving module 3, used for receiving the printing instruction from the user end in real time;

[0098] The parsing module 4 is used to parse the printing instruction and determine the printing requirements of the leaf vein image when receiving the printing instruction from the user terminal;

[0099] A screening module 5 is used to screen out leaf vein images that meet the printing requirements from the leaf vein images at multiple locations according to the printing requirements;

[0100] A fusion module 6 is used to fuse each filtered leaf vein image according to the screening result, and convert the fused leaf vein image into a vector image, wherein the leaf vein image fusion is used to synthesize the leaf vein image of the whole leaf;

[0101] The generating module 7 is used to generate a printing pattern file from the vector image.

[0102] As described above, it can be understood that the various components of the plant leaf vein custom identification printing system based on image recognition proposed in this application can realize the functions of any one of the plant leaf vein custom identification printing methods based on image recognition described above, and the specific structure will not be repeated.

[0103] Reference Figure 4 In an embodiment of the present application, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as monitoring data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a plant leaf vein custom identification printing method based on image recognition is implemented.

[0104] The processor executes the above-mentioned custom recognition and printing method of plant veins based on image recognition, including: acquiring a microscopic image of a leaf in real time, and determining whether the microscopic image meets the output conditions; when the microscopic image meets the output conditions, extracting a set of vein images in the microscopic image from multiple positions according to a preset acquisition standard; receiving a print instruction from a user terminal in real time; when receiving a print instruction from the user terminal, parsing the print instruction to determine the printing requirements of the vein image; according to the printing requirements, filtering out vein images that meet the printing requirements from the vein image sets at multiple positions; according to the filtering results, fusing each filtered vein image, and converting the fused vein image into a vector image, wherein the vein image fusion is used to synthesize the vein image of the entire leaf; generating a printing pattern file from the vector image.

[0105] The above-mentioned custom recognition and printing method of plant veins based on image recognition can obtain more comprehensive vein information by extracting vein images from multiple positions, ensuring that no important vein features are missed when synthesizing the vein image of the entire leaf. By acquiring high-quality microscopic images in real time, the high definition and details of the veins can be retained, so that the real vein texture can be reproduced when printing. The function of receiving the printing instructions of the user end in real time is provided, so that the user can customize the vein image according to his own needs, which increases the flexibility and user experience of the printing process. By screening and fusing the vein images of multiple positions, a complete and coherent vein image can be generated, which improves the accuracy and artistic value of the printed pattern. The fused vein image is converted into a vector image, which supports lossless enlargement and editing, ensuring the quality and customizability of the printed pattern. Through the automated processing flow, batch reproduction can be achieved, which improves the printing efficiency, while ensuring the consistency and accuracy of each printed pattern, and realizing batch printing of veins of the same leaf.

[0106] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, a method for customizing plant vein identification and printing based on image recognition is implemented, comprising the steps of: acquiring a microscopic image of a leaf in real time, and determining whether the microscopic image meets the output conditions; when the microscopic image meets the output conditions, extracting a set of leaf vein images in the microscopic image from multiple positions according to a preset acquisition standard; receiving a print instruction from a user end in real time; when the print instruction from the user end is received, parsing the print instruction to determine the printing requirements of the leaf vein image; according to the printing requirements, filtering out leaf vein images that meet the printing requirements from the leaf vein image sets at multiple positions; according to the filtering results, fusing each filtered leaf vein image, and converting the fused leaf vein image into a vector image, wherein the leaf vein image fusion is used to synthesize the leaf vein image of the entire leaf; generating a printing pattern file from the vector image.

[0107] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0108] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, device, article or method including the element.

[0109] The above description is only a preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for customizing plant vein identification and printing based on image recognition, characterized in that: The method comprises: Acquire a microscopic image of the leaf in real time, and determine whether the microscopic image meets the output conditions; When the microscopic image meets the output condition, extracting a leaf vein image set in the microscopic image from multiple positions according to a preset acquisition standard; Receive printing instructions from the user in real time; When receiving a printing instruction from the user terminal, parsing the printing instruction to determine the printing requirements of the leaf vein image; According to the printing requirement, selecting a leaf vein image that meets the printing requirement from a set of leaf vein images at multiple locations; According to the screening result, each screened leaf vein image is fused, and the fused leaf vein image is converted into a vector image, wherein the leaf vein image fusion is used to synthesize the leaf vein image of the whole leaf; The vector image is used to generate a printing pattern file.

2. The method for customizing plant vein identification and printing based on image recognition according to claim 1, characterized in that: Before the step of fusing each filtered leaf vein image, the method comprises: identifying each screened leaf vein image, and determining whether there is an image defect in the leaf vein image; When an image defect appears in the leaf vein image, marking the defect position in the leaf vein image, and analyzing the type of the defect appearing in the leaf vein image; When the defect type is that adjacent leaf veins are too close to each other, resulting in blurred images; Inputting the defect type and the leaf vein image into a preset segmentation model respectively, segmenting adjacent leaf veins through the segmentation model, and identifying target features of each leaf vein, wherein the target features include the branch point, texture, color and direction of the leaf vein; Filling the boundaries of the segmented leaf veins according to the target features through the segmentation model, and outputting the processed leaf vein image; Fusion is performed based on the processed leaf vein images.

3. The method for customizing plant vein identification and printing based on image recognition according to claim 2, characterized in that: After the steps of marking the defect position in the leaf vein image when an image defect appears in the leaf vein image, and analyzing the type of the defect appearing in the leaf vein image, the method further comprises: When the defect type is shadow and / or reflection on the leaf vein, identifying the illumination information of the defect position and the reflection characteristics of the leaf vein surface; Determining light radiation parameters on the surface of the leaf veins according to the illumination information and the reflection characteristics; Inputting the light radiation parameters and the light information into a preset light correction model, and generating light adjustment parameters through the light correction model; Segmenting the area where the leaf veins appear shadows and / or reflections, adjusting the illumination information based on illumination adjustment parameters, and re-correcting the leaf vein image at the defective position using the adjusted illumination information; Fusion is performed based on the corrected leaf vein images.

4. The method for customizing plant vein identification and printing based on image recognition according to claim 1, characterized in that: The method of fusing each filtered leaf vein image comprises: Obtain each filtered leaf vein image, and adapt the position of each leaf vein image; According to the adaptation results, mark the overlapping areas between different images; Identify the target features of the leaf veins in the overlapping area, construct an image coordinate system based on the target features, input the target features into the image coordinate system, and obtain a fusion position of the same overlapping area, wherein the target features include a branch point, texture, and direction of the leaf veins; The overlapping areas are merged according to the fusion positions, and the edges of the merged leaf veins are corroded and expanded; Based on the processing results, the fusion of leaf vein images is completed.

5. The method for customizing plant vein identification and printing based on image recognition according to claim 1, characterized in that: After the step of extracting the leaf vein image set in the microscopic image from multiple positions, the method further comprises: Calculating a differential image between the leaf vein image and an original microscopic image corresponding to the leaf vein image; According to the calculation result, obtaining the missing features in the difference image; Identifying whether the missing features in the differential image meet the condition of leaf vein features; Extracting the shape, texture and size information of the missing feature, inputting the shape, texture and size information of the missing feature and the leaf vein image into a preset matching model respectively, and analyzing the similarity coefficient between the missing feature and the leaf vein feature through the matching model; When the similarity coefficient is greater than a preset similarity threshold, it is determined that the missing feature meets the condition of the leaf vein feature; Based on the determination result, the missing features are filled into the leaf vein image to obtain an adjusted leaf vein image.

6. The method for customizing plant vein identification and printing based on image recognition according to claim 5, characterized in that: After the step of filling the missing features into the leaf vein image, the method further comprises: Determining whether the missing feature satisfies the condition of the broken feature of the leaf vein; Extracting edge features of the missing features, based on whether there are irregular fracture traces on the edges of the missing features; When there are irregular fracture marks on the edge of the missing feature, analyzing whether the texture feature of the missing feature is continuous with the target leaf vein; When the texture feature of the omitted feature is continuous with the target leaf vein, obtaining whether the cross-sectional width of the omitted feature matches the cross-sectional width of the target leaf vein; When the cross-sectional width of the omitted feature matches the cross-sectional width of the target leaf vein, analyzing whether the direction of the omitted feature is consistent with the direction of the target leaf vein; When the direction of the missing feature is consistent with the direction of the target leaf vein, it is determined that the missing feature meets the condition of the broken feature of the leaf vein; Based on the fact that the missing features satisfy the broken features of leaf veins, the missing features are filled onto the target leaf veins to obtain an adjusted leaf vein image.

7. The method for customizing plant vein identification and printing based on image recognition according to claim 1, characterized in that: The method for converting the fused leaf vein image into a vector image comprises: Identifying branch points of the fused leaf vein image, and dividing the leaf vein image into a plurality of regions according to the branch points; Extract the vein features of each segmented area; Convert each extracted leaf vein feature into a vector format to obtain a vector image block; Performing a first preprocessing on each vector image block, and synthesizing the first preprocessed vector image blocks; Based on the synthesis result, the vector image is subjected to a second preprocessing as a whole to obtain a final vector image.

8. A plant vein custom recognition and printing system based on image recognition, characterized in that: include: An acquisition module, used for acquiring a microscopic image of the leaf in real time and determining whether the microscopic image meets the output conditions; An extraction module, used for extracting a leaf vein image set in the microscopic image from multiple positions according to a preset acquisition standard when the microscopic image meets the output condition; A receiving module, used for receiving a printing instruction from a user terminal in real time; A parsing module, for parsing the printing instruction and determining the printing requirements of the leaf vein image when receiving the printing instruction from the user terminal; A screening module, used for screening leaf vein images that meet the printing requirements from leaf vein images at multiple locations according to the printing requirements; A fusion module, used for fusing each filtered leaf vein image according to the filtering result, and converting the fused leaf vein image into a vector image, wherein the leaf vein image fusion is used to synthesize the leaf vein image of the whole leaf; A generating module is used to generate a printing pattern file from the vector image.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images