Image processing system and method based on artificial intelligence

Abstract

The invention relates to the technical field of image processing, and particularly discloses an image processing system and method based on artificial intelligence, and the system comprises an image collection module, an image detection module, an image post-processing module, and an auxiliary decision module. The image detection module is used for classifying, identifying or detecting images by using an artificial intelligence technology and reducing the time required by data processing and target analysis, and the image post-processing module is used for analyzing, labeling or fusing detection results to generate information summaries for decision generation; the crop image is enhanced through the super-resolution technology, the pest information in the crop image is extracted through target detection, the pest control aid decision is generated according to analysis of the pest information, a user is helped to more accurately process the pest situation, and the pest control aid decision process is perfected through the data management technology and the weather forecast interface.

Description

Technical Field

[0001] The present invention belongs to the technical field of image processing, and in particular relates to an image processing system and method based on artificial intelligence. Background Art

[0002] In recent years, with the rapid development of deep learning technology and computer hardware equipment, great breakthroughs have been made in many fields such as image classification, object detection, and instance segmentation. Deep learning technology can automatically extract and represent features of raw image data. Among them, the popular convolutional neural network has a strong tolerance to image changes and strong fitting ability. Object detection and instance segmentation can be considered as a kind of target positioning, which is a key task in computer vision. It enables artificial intelligence systems to perceive and reason in an object-centric way.

[0003] The traditional way to identify agricultural pests and diseases is often for plant protection professionals or farmers to go to the field to observe the crop conditions and make judgments. The former has the problem of delayed feedback, while the latter is often judged by farmers based on their own experience, which may lead to increased pest resistance and excessive pesticide residues. There is no perfect crop image processing system in the industry today. They are either used for simple target detection tasks or give pesticide application suggestions by establishing dictionaries. They are not an integrated system that includes image detection and pesticide application suggestions. In addition, the insect images processed by the system must be obtained in accordance with certain standards to ensure correct species identification. However, in actual use, it may be difficult to obtain insect images according to standard requirements, or users will upload insect images of various standards through the Internet for automatic identification requests, so the widespread application of the system has certain difficulties and limitations.

[0004] Therefore, it is necessary to propose an image processing system and method based on artificial intelligence to at least partially solve the problems existing in the prior art.

[0005] The above information disclosed in this background technology is only used to increase the understanding of the background technology of the present invention and therefore, it may include information that does not constitute the prior art known to ordinary technicians in this field. Summary of the invention

[0006] The purpose of the present invention is to provide an image processing system and method based on artificial intelligence to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] An image processing system based on artificial intelligence comprises: an image acquisition module, an image detection module, an image post-processing module, an auxiliary decision module and a data management module. The image acquisition module is responsible for acquiring the captured field crop image data and analyzing the image quality. The image detection module classifies, identifies or detects the image using artificial intelligence technology to reduce the time required for data processing and target analysis. The image post-processing module analyzes, annotates or fuses the detection results to generate information summaries for use in decision making. The auxiliary decision module improves crop yield and quality by providing comprehensive analysis and data support and constructing corresponding suggestions based on crop diseases. The data management module is used to uniformly manage and store the received data to ensure the security of the image data and prevent unauthorized access.

[0009] Preferably, the data acquisition module periodically acquires photographic images of field crops, and records environmental data, crop varieties, and current growth status during each acquisition. At the same time, users can independently upload image data and fill in relevant information of field crops, and super-resolution the images using the LLAN network through the super-resolution port, and store the generated images in a database.

[0010] Preferably, the image detection module uses artificial intelligence technology to filter out pest images from image data. Here, the Faster RCNN network can be used to detect pests in the image, and then the pest area is segmented from the background through image segmentation and denoising, the pest feature values ​​are extracted to obtain the feature value matrix, and finally the classifier is used to automatically identify the pests.

[0011] In order to separate the pests from the background, the RGB components of the pests are first converted into HSV components, and then the H component image is segmented using the adaptive threshold solved by the maximum inter-class algorithm. The segmented image is binarized. After the pest image is segmented, its background image may still have isolated point noise. The maximum connected area is found for denoising, and finally the holes inside the maximum connected area are filled to obtain the final pest image.

[0012] The pest feature values ​​are extracted, and each set of feature values ​​can form a feature vector, as shown in the following formula:

[0013]

[0014] Where, ID i Is a numeric scalar used to represent the pest category, Index n is the eigenvalue index value that meets the classification of support vector machine, T nis the nth eigenvalue of the pest. For all the sample images of the pests to be identified, there is a two-dimensional matrix composed of X vectors. A support vector machine classifier is designed, and sample training is performed on the above two-dimensional eigenvalue matrix to obtain the classifier model. Then the system uses the classifier model to give automatic recognition results to the user's recognition request.

[0015] Preferably, the image post-processing module uses VGG ImageAnnotator to draw bounding boxes for detected pests, and labels pest categories according to the detection results, and then integrates similar detection results through a data fusion algorithm to provide more comprehensive and accurate pest information, and uses natural speech generation technology to convert the detection results into descriptive text, or present the data in the form of tables or graphs.

[0016] Preferably, the auxiliary decision module integrates data analysis results to provide dynamic decision support, including disease treatment plans, fertilization and irrigation recommendations, uses asynchronous processing technology to improve the system's response speed and processing capabilities, builds a rule-based expert system, and combines the knowledge of agricultural experts to generate treatment recommendations for different diseases.

[0017] Preferably, the data management module uses a MySQL database to store received data, archives historical data to a storage location that is less accessed to reduce the burden on the database, creates indexes based on category labels to achieve fast retrieval of image data, and uses table partitions to manage differently processed image data and related information to improve query performance, manages database connections through a connection pool, and reduces the overhead of connection creation and destruction.

[0018] An image processing method based on artificial intelligence, comprising:

[0019] Step 1: Obtain the captured field crop image or upload the image by the user, then save the image to the database and update the data list;

[0020] Step 2: Perform image super-resolution on the low-resolution image, and then apply artificial intelligence technology to perform a screening process based on pest features;

[0021] Step 3: Use a classifier to automatically identify pests based on pest images, process the output results, and generate information summaries;

[0022] Step 4: Use image processing to generate information for comprehensive analysis, combine historical data for global evaluation, and construct and output decision recommendations;

[0023] Step 5: Collect user feedback and suggestions, and regularly evaluate the system's detection accuracy to further improve system performance and user experience.

[0024] Preferably, in step 2, the processed low-resolution images are uniformly stored in a new file location, in step 3, the pest categories are marked according to the recognition results and similar pest images are integrated together, in step 4, decision suggestions can be extracted according to the answers of the large voice model Tongyi Qianwen, and in the face of serious diseases, "urgent" is marked in the decision suggestions, and an analysis report or notification information is generated based on the decision suggestions and sent to the user. The implementation time is generated simultaneously with the generation of the decision suggestions, and the weather conditions within a week are queried by calling the OpenWeatherMap weather API, and the nearest sunny or cloudy day is selected as the implementation time.

[0025] A terminal device, characterized in that it includes a processor and a computer-readable storage medium, the processor is used to implement various instructions; the computer-readable storage medium is used to store multiple instructions, and the instructions are suitable for being loaded by the processor and executed by any one of the above-mentioned artificial intelligence-based image processing systems.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] The present invention enhances crop images through super-resolution technology, uses target detection to extract pest information in crop images, generates auxiliary decision-making for pest control based on the analysis of pest information, helps users deal with pest situations more accurately, and uses data management technology and weather forecast interface to improve the auxiliary decision-making process for pests. In this process, the received images are preprocessed and feature extracted, pest feature values ​​are extracted to form feature vectors, and support vector machine classifiers are used for automatic identification, providing timely and accurate pest information for pest control, which is beneficial to reducing control costs and improving control effects.

[0028] The above summary is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments and features described above, further aspects, embodiments and features of the present invention will be readily apparent by reference to the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a framework diagram of an image processing system based on artificial intelligence of the present invention;

[0030] Figure 2 The figure is a flow chart of an image processing method based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. 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.

[0032] It should be further understood that the term “and / or” used in the specification of this application and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0033] Embodiment 1:

[0034] See also Figure 1 As shown, an image processing system based on artificial intelligence includes: an image acquisition module is responsible for acquiring the captured field crop image data and analyzing the image quality. It regularly acquires the captured images of the field crops. At the same time, users can upload image data independently and fill in the relevant information of the field crops. For low-resolution images, the images are super-resolved through the super-resolution port using the LLAN network, and the generated images are stored in the database.

[0035] The image detection module uses artificial intelligence technology to classify, identify or detect images, reducing the time required for data processing and target analysis. It uses artificial intelligence technology to filter out pest images from image data, and then uses image segmentation and denoising to separate the pest area from the background, extract pest feature values ​​to obtain a feature value matrix, and finally use a classifier to automatically identify the pests.

[0036] Pest feature value extraction selects two types of features of pests: one is based on the morphological features of the pests; the other is based on the color features of the pests.

[0037] (1) Morphological parameters of pests

[0038] ① Complexity

[0039] C=P 2 / 4πA

[0040]

[0041] P = A - SUM (in)

[0042] Where A and P represent the area and perimeter of the target respectively. The area is the total number of pixels occupied by the target in the image f(x, y). The size of the image is M×N, and SUM(in) is the total number of pixels in the 4-neighborhood that are all target points.

[0043] ②Duty cycle and elongation

[0044] B=A / L×W

[0045] S=W / L

[0046] Where L and W represent the length and width of the target minimum bounding rectangle, respectively.

[0047] ③Relative area and equivalent area circle radius

[0048] RA=A / SUM

[0049]

[0050] In the formula, SUM represents the total number of pixels in the entire image, and R represents the radius of the equivalent area circle.

[0051] ④Sphericity

[0052] SP=r i / r c

[0053] In the formula, r i Represents the radius of the target inscribed circle, r c Represents the radius of the target circumscribed circle, and the centers of the two circles are both at the center of gravity of the area.

[0054] (2) Color characteristic parameters of pests

[0055]

[0056] In the formula, h jk is the kth color component of the jth pixel in the image, G is the number of pixels in the image, and E 1 is the first-order moment, σ 2 is the second-order moment, ε 3 is the third-order moment.

[0057] The image post-processing module analyzes, annotates or fuses the detection results to generate information summaries for decision making. It uses VGG ImageAnnotator to draw bounding boxes for detected pests and annotate pest categories based on the detection results. It then uses a data fusion algorithm to integrate similar detection results to provide more comprehensive and accurate pest information. It uses natural speech generation technology to convert detection results into descriptive text or present data in the form of tables or graphs.

[0058] The decision support module constructs corresponding suggestions based on crop diseases through comprehensive analysis and data support to improve crop yield and quality. It integrates data analysis results, provides dynamic decision support, uses asynchronous processing technology to improve the system's response speed and processing capabilities, builds a rule-based expert system, and combines the knowledge of agricultural experts to generate treatment suggestions for different diseases.

[0059] The data management module is used to uniformly manage and store the received data, ensure the security of image data, and prevent unauthorized access. It uses the MySQL database to store the received data and archives historical data to a storage location with less access to reduce the burden on the database. It creates indexes based on category labels to achieve fast retrieval of image data, and uses table partitions to manage differently processed image data and related information to improve query performance. It manages database connections through a connection pool to reduce the overhead of connection creation and destruction.

[0060] Embodiment 2:

[0061] See also Figure 2 As shown, an image processing method based on artificial intelligence includes:

[0062] Step 1: Obtain the captured field crop image or upload the image by the user, then save the image to the database and update the data list;

[0063] Step 2: Perform image super-resolution on the low-resolution image, and then apply artificial intelligence technology to perform a screening process based on pest features;

[0064] Step 3: Use a classifier to automatically identify pests based on pest images, process the output results, and generate information summaries;

[0065] Step 4: Use image processing to generate information for comprehensive analysis, combine historical data for global evaluation, and construct and output decision recommendations;

[0066] Step 5: Collect user feedback and suggestions, and regularly evaluate the system's detection accuracy to further improve system performance and user experience.

[0067] In step 2, the processed low-resolution images are uniformly stored in a new file location. In step 3, the pest categories are labeled according to the recognition results and similar pest images are integrated together. In step 4, an analysis report or notification information is generated based on the decision recommendations and sent to the user.

[0068] In step one, the crop types, growth stages, and various environmental conditions are obtained, and the user adds the plots to which they belong, and can delete or modify the contents of the current data list; in step three, the eigenvalue matrix of the pests is extracted, a classifier is constructed using a support vector machine, and the classifier is trained using the obtained eigenvalue matrix. After the training is completed, this classifier can be used to automatically identify and classify pest identification requests.

[0069] As can be seen from the above, the present invention enhances crop images through super-resolution technology, uses target detection to extract pest information in crop images, generates auxiliary decision-making for pest control based on the analysis of pest information, helps users deal with pest situations more accurately, and uses data management technology and weather forecast interface to improve the auxiliary decision-making process for pests and diseases. In this process, the received images are preprocessed and feature extracted, pest feature values ​​are extracted to form feature vectors, and support vector machine classifiers are used for automatic identification, providing timely and accurate insect information for pest control, which is conducive to reducing control costs and improving control effects.

[0070] Embodiment 3:

[0071] An embodiment of the present invention also provides a terminal device, characterized in that it includes a processor and a computer-readable storage medium, the processor is used to implement various instructions; the computer-readable storage medium is used to store multiple instructions, and the instructions are suitable for being loaded by the processor and executing any one of the above-mentioned artificial intelligence-based image processing systems.

[0072] The embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, each process of the above-mentioned image processing system embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. Among them, the computer-readable storage medium, such as read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), magnetic disk or optical disk, etc.

[0073] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0074] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a program, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A way to specify functions in one or more boxes.

[0075] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0076] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0077] In the drawings of the embodiments disclosed in the present invention, only the structures involved in the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other.

[0078] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0079] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An image processing system based on artificial intelligence, comprising: The image acquisition module, the image detection module, the image post-processing module and the decision support module are characterized in that the image acquisition module is responsible for acquiring the captured field crop image data and analyzing the image quality; the image detection module uses artificial intelligence technology to classify, identify or detect images, reducing the time required for data processing and target analysis; the image post-processing module analyzes, annotates or fuses the detection results, generates information summaries for use in decision making; the decision support module improves crop yield and quality through comprehensive analysis and data support and constructs corresponding suggestions based on crop diseases.

2. The image processing system based on artificial intelligence according to claim 1, characterized in that: The data acquisition module regularly acquires photographic images of field crops. At the same time, users can upload image data independently and fill in relevant information of field crops. For low-resolution images, the images are super-resolved through the super-resolution port using the LLAN network, and the generated images are stored in the database.

3. The image processing system based on artificial intelligence according to claim 2, characterized in that: The image detection module uses artificial intelligence technology to filter out pest images from image data, then separates the pest area from the background through image segmentation and denoising, extracts pest feature values ​​to obtain a feature value matrix, and finally uses a classifier to automatically identify the pests.

4. The image processing system based on artificial intelligence according to claim 3, characterized in that: The image post-processing module uses VGG ImageAnnotator to draw bounding boxes for detected pests and annotate pest categories according to the detection results. It then integrates similar detection results through a data fusion algorithm to provide more comprehensive and accurate pest information, and uses natural speech generation technology to convert detection results into descriptive text or present data in the form of tables or graphs.

5. The image processing system based on artificial intelligence according to claim 4, characterized in that: The auxiliary decision-making module integrates data analysis results, provides dynamic decision support, uses asynchronous processing technology to improve the system's response speed and processing capabilities, builds a rule-based expert system, and combines the knowledge of agricultural experts to generate treatment suggestions for different diseases.

6. The image processing system based on artificial intelligence according to claim 5, characterized in that: The system also includes a data management module for uniformly managing and storing received data to ensure the security of image data and prevent unauthorized access. The data management module uses a MySQL database to store received data and archives historical data to a storage location with less access to reduce the burden on the database.

7. The image processing system based on artificial intelligence according to claim 6, characterized in that: The data management module creates indexes based on category labels to achieve rapid retrieval of image data, and uses table partitions to manage differently processed image data and related information to improve query performance, manages database connections through connection pools, and reduces the overhead of connection creation and destruction.

8. An image processing method based on artificial intelligence, based on an image processing system based on artificial intelligence according to any one of claims 1 to 7, characterized in that: include: Step 1: Obtain the captured field crop image or upload the image by the user, then save the image to the database and update the data list; Step 2: Perform image super-resolution on the low-resolution image, and then apply artificial intelligence technology to perform a screening process based on pest features; Step 3: Use a classifier to automatically identify pests based on pest images, process the output results, and generate information summaries; Step 4: Use image processing to generate information for comprehensive analysis, combine historical data for global evaluation, and construct and output decision recommendations; Step 5: Collect user feedback and suggestions, and regularly evaluate the system's detection accuracy to further improve system performance and user experience.

9. The image processing method based on artificial intelligence according to claim 8, characterized in that: In step 2, the processed low-resolution images are uniformly stored in a new file location. In step 3, the pest categories are marked according to the recognition results and similar pest images are integrated together. In step 4, an analysis report or notification information is generated according to the decision recommendations and sent to the user.

10. A terminal device, characterized in that: An artificial intelligence-based image processing system comprising a processor and a computer-readable storage medium, the processor being used to implement various instructions; the computer-readable storage medium being used to store multiple instructions, the instructions being suitable for being loaded by the processor and executed by any one of claims 1-7.

Images