Intelligent palm rattan identification and classification method and system

Abstract

The invention provides an intelligent rattan identification and classification method and system, and relates to the technical field of data processing, and the method comprises the steps: determining a rattan variety set, extracting image data of corresponding rattan according to the rattan variety set, outputting the image data as a sample image data set, carrying out the label labeling of the sample image data set, building a base identifier library, and carrying out the recognition and classification of the rattan. The method comprises the following steps: carrying out random extraction of a base identifier in a base identifier library based on a rattan variety set to obtain a plurality of identifier clusters, respectively carrying out supervised training on the plurality of identifier clusters, fusing the plurality of identifier clusters through an integrated learning method to obtain an identification classification model, and carrying out identification classification on rattan in a target scene through the identification classification model. The technical problems that in the prior art, a rattan identification and classification method depends on artificial experience and professional knowledge, so that the identification efficiency is low, and the accuracy is low are solved. The technical effects of intelligently identifying and classifying the rattan rattan and improving the rattan rattan type identification efficiency and accuracy are achieved.

Description

Technical Field

[0001] The invention relates to the technical field of data processing, and in particular to an intelligent palm vine identification and classification method and system. Background Art

[0002] Palm rattan occupies an important position in forestry, horticulture, medicinal materials and other fields due to its unique growth habits and wide application value. Palm rattan is not only an important economic crop, but its rattan and fruit also have important uses in home furnishings, handicrafts and medicinal materials. With people's attention to the palm rattan industry, how to effectively identify and classify different types of palm rattan has become an important research topic. However, there are many types of palm rattan, and the growth environment is complex, and there is often a large variability, which brings great challenges to identification and classification. Traditional palm rattan identification and classification methods mainly rely on manual experience and professional knowledge, which are inefficient and difficult to ensure the accuracy and consistency of identification.

[0003] The existing palm rattan identification and classification methods rely on manual experience and expertise, resulting in technical problems of low identification efficiency and low accuracy. Summary of the invention

[0004] The present application provides an intelligent palm vine identification and classification method and system, which is used to solve the technical problem that the palm vine identification and classification methods in the prior art rely on manual experience and professional knowledge, resulting in low identification efficiency and low accuracy.

[0005] In view of the above problems, the present application provides an intelligent palm vine identification and classification method and system.

[0006] The first aspect of the present application provides an intelligent palm rattan identification and classification method, the method comprising: interacting with a palm rattan knowledge base to determine a palm rattan species set, and extracting corresponding palm rattan image data based on the palm rattan species set, and outputting it as a sample image data set; labeling the sample image data set to obtain a labeled image data set; establishing a base identifier library, and randomly extracting base identifiers from the base identifier library based on the palm rattan species set to obtain multiple identifier clusters; using the labeled image data set as a sample space, respectively performing supervised training on multiple identifier clusters, and fusing the multiple identifier clusters through an ensemble learning method to obtain a recognition and classification model; and identifying and classifying palm rattan in a target scene through the recognition and classification model.

[0007] The second aspect of the present application provides an intelligent palm rattan identification and classification system, the system comprising: an image data extraction module, used to interact with a palm rattan knowledge base to determine a palm rattan species set, and extract corresponding palm rattan image data based on the palm rattan species set, and output it as a sample image data set; an image data annotation module, used to label the sample image data set and obtain an annotated image data set; a base identifier extraction module, used to establish a base identifier library, and randomly extract base identifiers from the base identifier library based on the palm rattan species set to obtain multiple identifier clusters; a classification model acquisition module, used to use the annotated image data set as a sample space, perform supervised training on multiple identifier clusters respectively, and fuse multiple identifier clusters through an ensemble learning method to obtain a recognition classification model; a data classification module, used to identify and classify palm rattan in a target scene through the recognition classification model.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] The method provided in the embodiment of the present application determines the set of palm rattan species through an interactive palm rattan knowledge base, and extracts the corresponding palm rattan image data according to the palm rattan species set, and outputs it as a sample image data set; performs label annotation of the sample image data set to obtain annotated image data set; establishes a base identifier library, and randomly extracts base identifiers based on the palm rattan species set in the base identifier library to obtain multiple identifier clusters; uses the annotated image data set as the sample space, performs supervised training of multiple identifier clusters respectively, and fuses multiple identifier clusters through an integrated learning method to obtain a recognition classification model; and performs recognition and classification of palm rattan in the target scene through the recognition classification model. The technical effect of intelligent recognition and classification of palm rattan and improving the efficiency and accuracy of palm rattan species recognition is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 A schematic diagram of the process of an intelligent palm vine identification and classification method provided in this application;

[0012] Figure 2 A schematic diagram of the structure of an intelligent palm vine identification and classification system provided in this application.

[0013] Explanation of the accompanying drawings: image data extraction module 11, image data labeling module 12, base identifier extraction module 13, classification model acquisition module 14, data classification module 15. DETAILED DESCRIPTION

[0014] The present application provides an intelligent palm rattan identification and classification method and system, which is used to solve the technical problem that the palm rattan identification and classification methods in the prior art rely on manual experience and professional knowledge, resulting in low identification efficiency and low accuracy. The technical effect of intelligent identification and classification of palm rattan and improving the efficiency and accuracy of palm rattan species identification is achieved.

[0015] Below, the technical solutions in the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments described herein. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the accompanying drawings, rather than all of them.

[0016] Embodiment 1, as Figure 1 As shown, the present application provides an intelligent palm vine identification and classification method, the method comprising:

[0017] The interactive palm rattan knowledge base determines a palm rattan species set, and extracts corresponding palm rattan image data according to the palm rattan species set, and outputs the data as a sample image data set.

[0018] Specifically, palm rattan is a vine plant belonging to the palm family, with a wide variety of species. Due to its unique growth habits and wide application value, it occupies an important position in many fields such as forestry, horticulture, and medicinal materials. The palm rattan species set of specific needs or conditions is determined by querying the palm rattan knowledge base through an interactive interface. The palm rattan knowledge base is a dynamic database that contains rich palm rattan related information, such as species name, morphological characteristics, ecological habits, geographical distribution, etc., and the palm rattan related information is verified and updated by experts to ensure its accuracy and reliability. The palm rattan species set includes a collection of different palm rattan species, varieties or cultivation types, such as industrial palm rattan, ornamental palm rattan, etc. According to the determined palm rattan species set, image data covering multiple conditions are extracted from the palm rattan knowledge base, and the image data include different growth stages such as seedling stage and mature stage, different shooting angles such as top and side, different lighting conditions such as strong light, weak light, and different growth environments such as wet and dry. The extracted image data are integrated and output as a sample image data set, which contains high-quality and comprehensive coverage of palm rattan images, providing a solid foundation and guarantee for intelligent palm rattan recognition and classification.

[0019] Furthermore, the palm rattan species set is determined by interacting with the palm rattan knowledge base, and the corresponding palm rattan image data is extracted according to the palm rattan species set, and output as a sample image data set, including: according to the scene environment information of the target scene, interacting with the palm rattan knowledge base to obtain the palm rattan species set; traversing the palm rattan species set to determine the stage characteristics of each type of palm rattan, and determining the corresponding stage sampling node set; taking the stage sampling node set and the palm rattan species as extraction constraints, extracting multiple palm rattan images based on big data, and storing them by classification to generate the sample image data set.

[0020] Specifically, the scene environment information of the target scene is obtained. The target scene refers to the place or environment where palm rattan identification is actually required. The scene environment information refers to various environmental variables related to the target scene, including geographical location, climatic conditions and industry needs. Climate conditions include temperature, humidity, precipitation, etc., and industry needs include market demand for palm rattan and cultivation mode. Through interaction with the palm rattan knowledge base and combined with the scene environment information of the target scene, the palm rattan species set that meets the scene is selected from the palm rattan knowledge base to ensure that the palm rattan data set can fully represent the diversity of palm rattan. A detailed analysis is performed on each palm rattan species set, and the species set is traversed to determine the growth stage characteristics of each palm rattan. For example, palm rattan may include different stages such as seedling stage, growth stage, and maturity stage, and each stage has different morphological characteristics and growth conditions. Stage characteristics refer to the characteristics of palm rattan at each stage in its life cycle, such as leaf morphology, rattan length, color change, etc. According to the stage characteristics of each palm rattan, the corresponding stage sampling node set is determined. The stage sampling node set refers to selecting appropriate time points for data sampling at different growth stages. Each node represents a specific sampling time point to ensure that effective and reliable palm rattan image data can be obtained at different growth stages. After determining the palm rattan species set and the stage sampling node set, the stage sampling node set and the palm rattan species are used as constraints. Based on big data technology, multiple images related to palm rattan are extracted from the palm rattan knowledge base. The extracted image data includes different growth stages, different angles, different lighting conditions and other situations of palm rattan, which can ensure the comprehensiveness of the image data. The extracted multiple palm rattan images are classified according to the species and growth stages of palm rattan, and effectively stored to generate a sample image data set, which provides comprehensive and reliable data support for intelligent palm rattan recognition and classification.

[0021] Perform label annotation on the sample image data set to obtain annotated image data set.

[0022] Furthermore, the sample image data set is labeled to obtain a labeled image data set, wherein the labels of the labeled image data set at least include palm rattan species, growth period status, maturity and disease status.

[0023] Specifically, after obtaining the sample image data set, image annotation tools such as LabelImg and CVAT are used according to image recognition technology, bioinformatics and other methods to add specific labels to each sample image in the sample image data set. The annotated labels include at least palm vine species, growth period status, maturity and disease conditions. The palm vine species refers to the palm vine species to which the annotated image belongs. The growth period status marks the growth status of the palm vine at the moment of the image, such as the seedling stage, growth stage, and maturity stage. The maturity index indicates whether the palm vine has reached full maturity or is in an immature state, such as the green fruit stage or the maturity stage. The disease condition refers to if the palm vine in the image is diseased, the type and degree of the disease are marked, such as yellow spot disease, leaf spot disease, etc. Optionally, the specific labeling process is to accurately determine the type of palm rattan through image recognition technology combined with the appearance characteristics of palm rattan, and label the type of palm rattan; analyze the size, color and other characteristics of the leaves and canes of the palm rattan in the image to determine which growth stage it is in, and label the generation period state; mark the maturity of the palm rattan fruit according to the size and color of the palm rattan fruit and the thickness of the cane in the image; identify whether the palm rattan leaves have spots or signs of diseases and pests through image analysis, and conduct qualitative and quantitative analysis of the diseases, and label the specific types of diseases and the degree of impact. For example, label leaf yellow spot disease, a minor disease. Labeling the sample image data set provides accurate and comprehensive data support for intelligent palm rattan recognition, thereby improving the efficiency and accuracy of palm rattan recognition.

[0024] A base identifier library is established, and base identifiers are randomly extracted from the base identifier library based on the palm rattan species set to obtain multiple identifier clusters.

[0025] Furthermore, a base identifier library is established, and base identifiers are randomly extracted from the base identifier library based on the palm rattan species set to obtain multiple identifier clusters, including: defining an available model structure set, and establishing multiple base identifiers based on the available model structure set to obtain the base identifier library; obtaining multidimensional feature information corresponding to the palm rattan species set for heavy evaluation, and defining an adjustment coefficient set based on the heavy evaluation result; based on a preset cluster size and the adjustment coefficient set, repeatable random extraction is performed in the base identifier library, and the extracted content is output as multiple identifier clusters.

[0026] Specifically, a set of available model structure sets is first determined, and the available model structure set is a verified model architecture, which includes a variety of machine learning models suitable for palm rattan identification, such as convolutional neural networks, deep neural networks, and transfer learning models, representing different network structures and parameter configurations, and can capture different features in palm rattan images. According to the available model structure set, multiple base identifiers are established, each of which is initialized with a model structure in the set and trained with annotated image data sets to perform specific palm rattan species or feature classification tasks. Multiple base identifiers are integrated to form a library containing multiple base identifiers. Multidimensional feature information corresponding to the palm rattan species set is extracted from the palm rattan image through a feature extraction algorithm, including palm rattan intrinsic features, value features, and typical planting ratios, such as intrinsic features such as leaf morphology and rattan length, value features such as economic value and market demand, and typical planting ratios: such as the distribution of different palm rattan species in the target scene. The extracted multidimensional feature information is cleaned, normalized or standardized to ensure data consistency, and then the correlation coefficient or mutual information between each feature and the palm rattan species is calculated to perform a heavy evaluation on the multidimensional feature information. The heavy evaluation refers to a weighted calculation based on the multidimensional features, which is used to judge the importance of different palm rattan species. According to the analysis results, a heavy evaluation result is obtained. Then, based on the heavy evaluation result, an adjustment coefficient set is defined. The adjustment coefficient set is a parameter determined according to the evaluation result, which is used to adjust the randomness and weight of the model selection, representing the range of up and down fluctuations. The preset cluster size is determined. The cluster size refers to the number of base identifiers contained in each identifier cluster. The identifier cluster is a set of multiple base identifiers, and each cluster has unique classification capabilities and combination characteristics. For example, an identifier cluster contains M base identifiers, where M is a positive integer. According to the cluster size and the adjustment coefficient set, random extraction is performed in the base identifier library. The random extraction adopts a put-back strategy, that is, after one extraction, the base identifier can still participate in the next extraction to ensure diversity. Through multiple extractions, multiple identifier clusters are formed, and each cluster is used for subsequent classification model training and integrated learning. By constructing a base identifier library, a rich model diversity is provided, which helps to improve the performance of intelligent palm vine identification. By heavy evaluation and adjustment of the coefficient set, it is ensured that the importance of features is taken into account in the random extraction process, making the extracted identifier clusters more in line with actual needs, further improving the accuracy and stability of palm vine identification.

[0027] Furthermore, multidimensional feature information corresponding to the palm rattan species set is obtained for heavy evaluation, and an adjustment coefficient set is defined based on the heavy evaluation result, including: interacting with the palm rattan knowledge base to match the intrinsic feature information corresponding to the palm rattan species set; interacting with the palm rattan trading platform to obtain the value feature information corresponding to the palm rattan species set; interacting with the target scene to obtain the typical planting ratio information corresponding to the palm rattan species set; performing weighted evaluation based on the intrinsic feature information, the value feature information and the typical planting ratio information to obtain the heavy evaluation result; calculating the average value of the heavy evaluation result as the heavy baseline, and calculating the ratio of the heavy evaluation result to the heavy baseline, and outputting it as the adjustment coefficient set.

[0028] Specifically, the intrinsic feature information of the palm rattan species set is obtained through the interactive palm rattan knowledge base. The intrinsic feature information refers to the natural attributes of the palm rattan, which is used to reflect the difficulty of identification, such as rattan length, leaf shape and size, morphological complexity, texture significance, etc. These features directly reflect the biological characteristics of the palm rattan. Through the interactive palm rattan trading platform, that is, through the interactive online or offline palm rattan trading market, the value feature information corresponding to the palm rattan species set is obtained. The value feature refers to information reflecting market demand and economic benefits. For example, a certain type of palm rattan has a high price due to its strong durability. By analyzing the actual target area or environment, the typical planting ratio information of the palm rattan species in the target scene is extracted. The typical planting ratio information refers to the planting area ratio or quantity distribution of various types of palm rattan in a specific area, reflecting the actual needs of the target scene. According to the task requirements and actual importance, each feature device weight is used, and each palm rattan species is comprehensively evaluated using a weighted formula. The weighted evaluation refers to assigning weights according to the importance of the features, calculating the comprehensive score, and obtaining a heavy evaluation result. The result of the weight assessment = (intrinsic characteristic value × w1) + (value characteristic value × w2) + (planting ratio characteristic value × w3), where w1, w2 and w3 are weights. The weight assessment results of all species are averaged and the average calculation result is used as the weight baseline. The weight baseline represents the average level of importance of all species and is the reference value for calculating the adjustment coefficient. The weight assessment score of each palm rattan species is compared with the weight baseline to obtain the adjustment coefficient set, where the adjustment coefficient = weight assessment score / weight baseline, that is, the adjustment coefficient is a floating ratio based on the importance of the species relative to the average level, which is used to dynamically adjust the probability or number of identifier selection. By combining the intrinsic characteristics, value characteristics and planting ratio characteristics, a comprehensive weight assessment is formed, which can dynamically adapt the configuration of different types of identifiers according to actual needs. Through the dynamic adjustment mechanism of the adjustment coefficient, the flexibility and adaptability of palm rattan identification are enhanced, thereby improving the efficiency and accuracy of palm rattan classification.

[0029] Taking the labeled image data set as the sample space, supervised training of the plurality of the identifier clusters is performed respectively, and the plurality of the identifier clusters are fused through an ensemble learning method to obtain a recognition classification model.

[0030] The palm vines in the target scene are identified and classified by the identification and classification model.

[0031] Specifically, the labeled image dataset is used as the sample space. The sample space refers to the training data used to train the model. The labeled image dataset has been accurately labeled and contains label information such as the type, growth period, maturity, and disease condition of palm rattan. Using the labeled image dataset, supervised training is performed for each recognizer cluster. Supervised training refers to training the model on the labeled sample data, optimizing the model parameters by inputting image data and corresponding labels, so that the model can recognize the relationship between the image and the label. Each recognizer cluster contains multiple base recognizers, which are responsible for classifying different palm rattan species, stages or disease conditions. The training process is to use the sample image dataset as input for each recognizer cluster. The sample image dataset contains input images, such as palm rattan images and corresponding labels. The training goal is to learn the relationship between the input image and the label so that the base recognizer can accurately predict the corresponding label on the new input image. Through traditional supervised learning method training, each base recognizer extracts features from the sample space, and uses the cross entropy loss function to measure the gap between the model prediction and the true label. The base recognizer calculates the prediction result through forward propagation, and then calculates the gradient through back propagation according to the result of the loss function. The model weight is adjusted using optimization algorithms such as SGD and Adam. When the model accuracy reaches the convergence state, the iteration is stopped to obtain the trained recognizer cluster. After training, each recognizer cluster will learn the best weight parameters and can accurately classify the input image. Through ensemble learning methods such as voting method and weighted average method, the trained multiple recognizer clusters are fused to generate the final overall recognition and classification model. The recognition and classification model reduces the deviation and variance of a single model through the combination of multiple models, and improves the performance of the overall model. During the fusion process, the weight can be dynamically adjusted according to the accuracy of each recognizer cluster, so that the overall model optimizes the recognition effect of each palm rattan species, stage and disease condition. By integrating multiple recognizer clusters, the possible bias and overfitting problems of a single model are reduced, and multiple tasks can be processed simultaneously, thereby improving the overall processing capability and task adaptability, and providing accurate, efficient and stable palm vine classification and recognition functions.

[0032] Furthermore, the image data in the target scene is collected through shooting, sensor data or existing databases, and the palm vine image in the target scene is passed as input data to the trained recognition and classification model. The recognition and classification model includes multiple base identifier clusters. The outputs of multiple base identifiers are fused through an integrated learning method to obtain the final classification result. Multiple base identifier clusters in the recognition and classification model respectively identify and classify the input image and fuse them, and output specific category information of the palm vine in the target scene, including palm vine species, growth stage, maturity, and whether there is disease. Through the integrated learning method, multiple base identifier clusters can complement each other and improve the classification accuracy of palm vine species and growth stages. Through automated image recognition and classification, the recognition and classification model can efficiently and accurately classify and identify the palm vine in the target scene, greatly reducing the cost and time of manual inspection.

[0033] Furthermore, label annotation is performed on the sample image dataset to obtain the annotated image dataset, and then the method further includes: performing image enhancement on the annotated image dataset to generate an enhanced image dataset; extracting palm vine edge information through edge detection; performing image segmentation based on the edge information and identifying and determining the palm vine pixel area; and performing lightweight processing on the annotated image dataset based on the palm vine pixel area.

[0034] Specifically, the labeled image dataset is enhanced by a variety of data enhancement methods to generate an enhanced image dataset. Data enhancement methods include geometric transformation, color transformation, etc. Geometric transformation includes rotation, flipping, cropping and scaling to simulate palm vine images at different shooting angles and distances. Color transformation such as changing brightness, contrast, saturation, etc. simulates different lighting conditions. The image enhancement refers to expanding the scale and diversity of the training dataset by transforming and processing the image. The enhanced image dataset refers to a new image set generated by enhancement, which is more diverse and representative. The Canny edge detection algorithm, Sobel operator or Laplace operator are used to identify the areas in the image where the pixel values ​​change significantly to perform palm vine image edge detection and extract palm vine edge information. Then, according to the results of edge detection, the image is divided into a target area and a background area. The target area is the palm vine pixel area. The image segmentation method is optional. Starting from the initial seed point, the area is gradually expanded according to the similarity of adjacent pixels through the region growing method, or the palm vine area is accurately segmented using a semantic segmentation model. After determining the pixel area of ​​palm vines, the labeled image dataset was lightened to simplify the image content, retaining only the pixel area related to palm vines, reducing noise interference and improving model training efficiency. The generated enhanced image dataset and the lightened labeled image dataset significantly improved the quality of training data and the accuracy and reliability of model recognition.

[0035] Furthermore, the light-weight processing of the annotated image dataset is performed based on the palm vine pixel area, including: extracting pixel coordinate features of the palm vine pixel area; mapping the pixel coordinate features to the annotated image dataset, and eliminating pixels that do not belong to the mapping results to achieve light-weight processing.

[0036] Specifically, after image segmentation, the divided palm vine pixel area is obtained, and then the segmented image is traversed to extract the pixel coordinates of the palm vine pixel area to obtain the pixel coordinate features of the palm vine pixel area. The palm vine pixel area refers to a group of continuous pixels in the palm vine image, indicating the location of the target object. The pixel coordinate feature refers to the position coordinate of each target pixel in the image, represented by (x, y). The extracted palm vine pixel area coordinate features are mapped to the annotated image dataset, and each image in the annotated dataset is checked one by one to ensure that the target area is completely aligned with the pixel coordinates. The target area is reconstructed through the coordinate features, and only the pixels matching the pixel coordinates are retained in the annotated image dataset. After the mapping is completed, the pixels in the image that do not belong to the target area, that is, the background area or other interference parts, are removed, and the non-target area pixels are set to specific values, such as zero or completely removed, so as to simplify the image structure and realize the lightweight processing of the annotated image dataset. Through lightweight processing, each image only retains the pixel area related to the palm vine, so that the model can focus more on the target features. By accurately extracting pixels in the target area and removing the background, the interference of irrelevant information on model training is significantly reduced, allowing the model to focus on learning the key features of the target area and improve the accuracy of recognition and classification.

[0037] In summary, the intelligent palm vine identification and classification method provided in the embodiment of the present application has the following technical effects:

[0038] 1. Through the intelligent palm vine identification and classification method, palm vines can be identified automatically and quickly, which significantly improves the recognition efficiency and accuracy and reduces the cost of manual identification.

[0039] 2. Accurate identification and classification will help promote the development of palm rattan breeding, cultivation, processing and other industries, and improve the economic value and ecological benefits of palm rattan resources.

[0040] Embodiment 2, based on the same inventive concept as the intelligent palm vine identification and classification method in the above embodiment, Figure 2 As shown, the present application provides an intelligent palm rattan identification and classification system, wherein the system comprises:

[0041] An image data extraction module 11 is used to interact with a palm rattan knowledge base to determine a palm rattan species set, and extract the corresponding palm rattan image data according to the palm rattan species set, and output it as a sample image data set; an image data annotation module 12 is used to label the sample image data set and obtain an annotated image data set; a base identifier extraction module 13 is used to establish a base identifier library, and randomly extract base identifiers from the base identifier library based on the palm rattan species set to obtain multiple identifier clusters; a classification model acquisition module 14 is used to use the annotated image data set as a sample space, perform supervised training on multiple identifier clusters respectively, and fuse multiple identifier clusters through an ensemble learning method to obtain a recognition classification model; a data classification module 15 is used to identify and classify palm rattan in a target scene through the recognition classification model.

[0042] Furthermore, the image data extraction module 11 is also used to perform the following steps: according to the scene environment information of the target scene, interact with the palm rattan knowledge base to obtain the palm rattan species set; traverse the palm rattan species set to determine the stage characteristics of each type of palm rattan, and determine the corresponding stage sampling node set; use the stage sampling node set and the palm rattan species as extraction constraints to extract multiple palm rattan images based on big data, and classify and store them to generate the sample image data set.

[0043] Furthermore, the base identifier extraction module 13 is also used to perform the following steps: define an available model structure set, and establish multiple base identifiers based on the available model structure set to obtain the base identifier library; obtain the multidimensional feature information corresponding to the palm rattan species set for heavy evaluation, and define an adjustment coefficient set based on the heavy evaluation result; based on the preset cluster size and the adjustment coefficient set, perform repeatable random extraction in the base identifier library, and output the extracted content as multiple identifier clusters.

[0044] Furthermore, the base identifier extraction module 13 is also used to perform the following steps: interact with the palm rattan knowledge base to match the intrinsic feature information corresponding to the palm rattan species set; interact with the palm rattan trading platform to obtain the value feature information corresponding to the palm rattan species set; interact with the target scene to obtain the typical planting ratio information corresponding to the palm rattan species set; perform weighted evaluation based on the intrinsic feature information, the value feature information and the typical planting ratio information to obtain the heavy assessment result; calculate the average value of the heavy assessment result as the heavy baseline, and calculate the ratio of the heavy assessment result to the heavy baseline, and output it as the adjustment coefficient set.

[0045] Furthermore, the image data annotation module 12 is also used to perform the following steps: the labels of the annotated image data set at least include palm rattan species, growth period status, maturity and disease status.

[0046] Furthermore, the image data annotation module 12 is also used to perform the following steps: perform image enhancement on the annotated image data set to generate an enhanced image data set; extract palm vine edge information through edge detection; perform image segmentation based on the edge information and identify and determine the palm vine pixel area; and perform lightweight processing on the annotated image data set based on the palm vine pixel area.

[0047] Furthermore, the image data annotation module 12 is also used to perform the following steps: extracting pixel coordinate features of the palm vine pixel area; mapping the pixel coordinate features to the annotated image data set, and removing pixels that do not belong to the mapping results to achieve lightweight processing.

[0048] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0049] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. An intelligent palm vine identification and classification method, characterized in that: The method comprises: The interactive palm rattan knowledge base determines a palm rattan species set, and extracts corresponding palm rattan image data according to the palm rattan species set, and outputs the data as a sample image data set; Perform label annotation on the sample image data set to obtain annotated image data set; Establishing a base identifier library, and randomly extracting base identifiers based on the palm rattan species set in the base identifier library to obtain a plurality of identifier clusters; Taking the labeled image data set as a sample space, respectively performing supervised training on a plurality of the identifier clusters, and fusing the plurality of the identifier clusters through an ensemble learning method to obtain a recognition classification model; The palm vines in the target scene are identified and classified by the identification and classification model.

2. An intelligent palm vine identification and classification method as claimed in claim 1, characterized in that: The interactive palm rattan knowledge base determines a palm rattan species set, and extracts corresponding palm rattan image data according to the palm rattan species set, and outputs the data as a sample image data set, including: According to the scene environment information of the target scene, interact with the palm rattan knowledge base to obtain the palm rattan species set; Traversing the palm rattan species set to determine the stage characteristics of each type of palm rattan, and determining the corresponding stage sampling node set; Taking the stage sampling node set and the palm vine species as extraction constraints, a plurality of palm vine images based on big data are extracted and classified and stored to generate the sample image data set.

3. An intelligent palm rattan identification and classification method as claimed in claim 2, characterized in that: A base identifier library is established, and base identifiers are randomly extracted from the base identifier library based on the palm rattan species set to obtain multiple identifier clusters, including: Defining an available model structure set, and establishing a plurality of base identifiers based on the available model structure set, and obtaining the base identifier library; Acquiring multidimensional feature information corresponding to the palm rattan species set for heavy evaluation, and defining an adjustment coefficient set based on the heavy evaluation result; Based on the preset cluster size and the adjustment coefficient set, repeatable random extraction is performed in the base identifier library, and the extracted content is output as a plurality of the identifier clusters.

4. An intelligent palm rattan identification and classification method as claimed in claim 3, characterized in that: The multidimensional feature information corresponding to the palm rattan species set is obtained for heavy evaluation, and an adjustment coefficient set is defined based on the heavy evaluation result, including: Interacting with the palm rattan knowledge base to match the intrinsic feature information corresponding to the palm rattan species set; Interacting with the palm rattan trading platform to obtain value characteristic information corresponding to the palm rattan species set; Interact with the target scene to obtain typical planting ratio information corresponding to the palm rattan species set; Performing a weighted evaluation based on the intrinsic feature information, the value feature information and the typical planting ratio information to obtain the heavy assessment result; An average value of the severe assessment results is calculated as a severe baseline, and a ratio of the severe assessment results to the severe baseline is calculated and output as the adjustment coefficient set.

5. The intelligent palm vine identification and classification method according to claim 1, characterized in that: The sample image data set is labeled to obtain a labeled image data set, wherein the labels of the labeled image data set at least include palm rattan species, growth period status, maturity and disease status.

6. The intelligent palm rattan identification and classification method according to claim 1, characterized in that: Label the sample image data set to obtain the labeled image data set, and then further include: Perform image enhancement on the annotated image dataset to generate an enhanced image dataset; Through edge detection, the edge information of palm vine is extracted; Perform image segmentation based on edge information and identify and determine the palm vine pixel area; The labeled image data set is lightly processed based on the palm vine pixel area.

7. An intelligent palm rattan identification and classification method as claimed in claim 6, characterized in that: The light processing of the annotated image data set is performed based on the palm vine pixel area, including: Extracting pixel coordinate features of the palm vine pixel area; The pixel coordinate features are mapped to the annotated image data set, and pixels that do not belong to the mapping results are removed to achieve lightweight processing.

8. An intelligent palm rattan identification and classification system, characterized in that: The intelligent palm rattan identification and classification system is used to implement the steps of any one of the methods described in claims 1 to 7, including: An image data extraction module is used to interact with a palm rattan knowledge base to determine a palm rattan species set, and extract corresponding palm rattan image data according to the palm rattan species set, and output it as a sample image data set; An image data annotation module is used to annotate the sample image data set with labels to obtain annotated image data set; A base identifier extraction module is used to establish a base identifier library, and randomly extract base identifiers based on the palm rattan species set in the base identifier library to obtain multiple identifier clusters; A classification model acquisition module is used to perform supervised training of the plurality of the identifier clusters respectively using the annotated image data set as a sample space, and fuse the plurality of the identifier clusters through an ensemble learning method to obtain a recognition classification model; The data classification module is used to identify and classify palm vines in the target scene through the identification and classification model.

Images