A human-computer interaction optimization system and method based on image processing

By preprocessing and feature extraction of interactive images, iterative optimization combined with user feedback, and building a benchmark image database, the problems of low efficiency and poor user experience in traditional image processing technology are solved, and fast and accurate image processing and Efficient user interaction.

CN118445438BActive Publication Date: 2024-11-05ROCO YINZHOU (SHENZHEN) TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410617784.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-17
Publication Date
2024-11-05
Estimated Expiration
2044-05-17

AI Technical Summary

Technical Problem

Traditional image processing techniques have problems of inefficiency and poor user experience when processing interactive images, especially when user needs are not comprehensive and delays are increased during the interaction process.

Method used

By acquiring interactive images for preprocessing, user needs are extracted to separate related features and non-associated features, the images are optimized, and iterated in real time until the user is satisfied. At the same time, user needs are summarized to build a benchmark image database to quickly match user needs.

Benefits of technology

Improves the efficiency and quality of image processing, reduces output delay, and improves user experience and interaction efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118445438B_ABST
    Figure CN118445438B_ABST
Patent Text Reader

Abstract

The present invention belongs to the field of image processing technology, and specifically relates to a human-computer interaction optimization system and method based on image processing. The invention obtains a first reference image by preprocessing the interactive image, and then extracts features of the first reference image according to user needs to obtain associated features and non-associated features. Then, the first reference image is optimized according to the associated features and non-associated features to obtain a second reference image, and the second reference image is iteratively optimized according to user feedback and new requirements until the user is satisfied. At the same time, the present invention also pairs the second reference image with user requirements to form a reference image database. In this way, the most matching paired image can be quickly found according to different user needs, thereby improving the efficiency and accuracy of human-computer interaction and reducing the output delay of image processing.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of image processing, and in particular relates to a human-computer interaction optimization system and method based on image processing. Background Art

[0002] In the field of digital image processing, the processing and application of interactive images are receiving more and more attention. Interactive images refer to the process of interaction between users and systems through images in computer vision systems. Such interactive image processing has important application value in many fields, such as unmanned driving, smart medical care, virtual reality, etc. With the continuous development of artificial intelligence and computer vision technology, the application of human-computer interaction in the field of image processing is becoming more and more extensive. In order to meet the needs of users for interactive image processing, improving processing efficiency and quality is an urgent problem to be solved.

[0003] However, traditional image processing technology often faces some challenges when processing interactive images. When processing images input by users, the images are often processed comprehensively. However, the actual needs of users may not involve the entire image. Comprehensive processing of images will obviously lead to increased output delays. In addition, in the process of interacting with users, users may also put forward some new requirements. The high delay of comprehensive image processing will obviously lead to a decrease in user experience. To solve the above problems, the present invention provides a human-computer interaction optimization method based on image processing. Summary of the invention

[0004] The purpose of the present invention is to provide a human-computer interaction optimization system and method based on image processing, which can accurately and quickly process interactive images according to user needs, improve processing efficiency and quality, and enhance user experience.

[0005] The technical solution adopted by the present invention is as follows:

[0006] A human-computer interaction optimization method based on image processing, comprising:

[0007] Acquire an interactive image, and preprocess the interactive image to obtain a first reference image;

[0008] Obtaining user needs, and performing feature extraction on the first reference image according to the user needs to obtain associated features and non-associated features;

[0009] Optimizing the first reference image according to the associated features and the non-associated features, and calibrating the optimized first reference image as the second reference image;

[0010] The second reference image is displayed to the user, and user feedback and new requirements are collected in real time, and the second reference image is iteratively optimized according to the user feedback and new requirements until the user no longer provides feedback and new requirements;

[0011] The newly added requirements and user requirements are aggregated into independent requirements, the second reference images iterated under the newly added requirements and user requirements are calibrated as paired images corresponding to the independent requirements, and the independent requirements and paired images are aggregated into a reference image database.

[0012] In a preferred embodiment, the step of acquiring the interactive image and preprocessing the interactive image to obtain the first reference image includes:

[0013] Acquire the interactive image, and perform denoising and enhancement processing on the interactive image;

[0014] Resize and color correct the interaction image after denoising and enhancement to obtain a front image;

[0015] The front image is partitioned and classified into a plurality of block images, which are marked as first reference images.

[0016] In a preferred solution, the step of performing partition processing on the front image to classify the front image into a plurality of block images includes:

[0017] Acquire size information and color information of the front image;

[0018] According to the size information, the front image is equally divided into a number of regions of the same size and marked as initial region images;

[0019] Performing color cluster analysis on the initial region images according to the color information to obtain color distribution characteristics of each initial region image;

[0020] According to the color distribution characteristics, initial region images with similar color distributions in adjacent initial region images are merged to obtain a plurality of block images.

[0021] In a preferred embodiment, the step of performing color cluster analysis on the initial region images to obtain color distribution characteristics of each initial region image includes:

[0022] Obtaining the color value of each pixel in the initial area image;

[0023] The number of clusters is obtained and marked as a reference parameter, and a color value consistent with the number of the reference parameter is randomly selected as a cluster center;

[0024] Counting the frequency of each color cluster center appearing in the initial region image and marking it as the first condition parameter;

[0025] Obtaining a screening threshold, and comparing the screening threshold with a first condition parameter;

[0026] If the first condition parameter is greater than the screening threshold, the color corresponding to the first condition parameter is marked as the main color, and the color distribution characteristics of the initial area image are generated according to the main color;

[0027] If the first condition parameter is less than or equal to the screening threshold, the corresponding color is marked as a secondary color and is not recorded in the color distribution feature.

[0028] In a preferred embodiment, the step of merging the initial region images with similar color distribution in adjacent initial region images according to the color distribution characteristics comprises:

[0029] Acquire the color distribution characteristics of the adjacent initial region images, and synchronously perform vectorization processing to obtain the second conditional parameter;

[0030] Obtaining a standard function, and inputting the second condition parameter into the standard function, and marking the output result as an evaluation parameter;

[0031] Obtaining an evaluation threshold, and comparing the evaluation threshold with an evaluation parameter;

[0032] If the evaluation parameter is greater than the evaluation threshold, the corresponding adjacent initial area images are merged to obtain a merged block image;

[0033] If the evaluation parameter is less than or equal to the evaluation threshold, a splicing curve is constructed at the connection between the corresponding adjacent initial region images.

[0034] In a preferred embodiment, the step of obtaining user requirements and extracting features from the first reference image according to the user requirements to obtain associated features and non-associated features includes:

[0035] Collecting voice information or text information corresponding to the user's needs, wherein the voice information will be synchronously converted into text information;

[0036] Extracting a block image corresponding to the user's needs from the first reference image, and performing an offset process on the edge curve of the block image to obtain a feature extraction curve;

[0037] Image segmentation is performed on the first reference image according to the feature extraction curve, and the block image within the feature extraction curve is marked as the associated feature, and the block image outside the feature extraction curve is marked as the non-associated feature.

[0038] In a preferred embodiment, the step of iteratively optimizing the second reference image according to the user feedback and the new requirements until the user no longer provides feedback and new requirements comprises:

[0039] Displaying the second reference image to the user, and collecting the user's feedback and new requirements in real time;

[0040] If the user does not send any feedback or additional requirements, the second reference image is recorded as the standard image and added to the reference image database;

[0041] If the user issues feedback and new requirements, synchronously optimizing the second reference image according to the user's feedback and new requirements, including adjusting image parameters, size and content, and outputting the optimized image;

[0042] Collect user feedback and new requirements under the optimized image, and stop the optimization when the user no longer provides feedback and new requirements. Otherwise, continue to output the optimized image until the user no longer provides feedback and new requirements, and record the optimized image under each new requirement and add it to the benchmark image database.

[0043] In a preferred embodiment, the step of aggregating the independent requirements and paired images into a reference image database includes:

[0044] Classify and sort out the newly added requirements and user requirements, group the same or similar requirements into one category, and obtain an independent requirement list;

[0045] Matching each independent requirement in the independent requirement list with the second reference image to find a paired image that best matches the independent requirement;

[0046] The paired images are associated with corresponding independent requirements and summarized into a reference image database.

[0047] The present invention also provides a human-computer interaction optimization system based on image processing, which is applied to the above-mentioned human-computer interaction optimization method based on image processing, comprising:

[0048] An image acquisition module, the image acquisition module is used to acquire an interactive image and preprocess the interactive image to obtain a first reference image;

[0049] A feature extraction module, the feature extraction module is used to obtain user requirements, and extract features from the first reference image according to the user requirements to obtain associated features and non-associated features;

[0050] An optimization module, the optimization module is used to optimize the first reference image according to the associated features and the non-associated features, and calibrate the first reference image after the optimization as the second reference image;

[0051] an iteration module, the iteration module being used to display the second reference image to the user, collect user feedback and new requirements in real time, and iteratively optimize the second reference image according to the user feedback and new requirements until the user no longer issues feedback and new requirements;

[0052] A database construction module, the database construction module is used to aggregate the new requirements and user requirements into independent requirements, calibrate the second reference images iterated under the new requirements and user requirements as paired images corresponding to the independent requirements, and aggregate the independent requirements and paired images into a reference image database.

[0053] And, a human-computer interaction optimization device based on image processing, comprising:

[0054] at least one processor;

[0055] and a memory communicatively coupled to the at least one processor;

[0056] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned human-computer interaction optimization method based on image processing.

[0057] The technical effects achieved by the present invention are:

[0058] The present invention obtains a first reference image by preprocessing the interactive image, then extracts features from the first reference image according to user needs to obtain associated features and non-associated features, then optimizes the first reference image according to the associated features and non-associated features to obtain a second reference image, and iteratively optimizes the second reference image according to user feedback and new needs until the user is satisfied. At the same time, the present invention also pairs the second reference image with user needs to form a reference image database. In this way, the best matching paired image can be quickly found according to different user needs, thereby improving the efficiency and accuracy of human-computer interaction and reducing the output delay of image processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a flow chart of the method of the present invention;

[0060] Figure 2 It is a system module diagram of the present invention;

[0061] Figure 3 It is a structural diagram of the device of the present invention. DETAILED DESCRIPTION

[0062] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0063] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0064] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention. The phrase "in a preferred embodiment" that appears in different places in this specification does not refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0065] See also Figure 1 As shown, the present invention provides a human-computer interaction optimization method based on image processing, comprising:

[0066] S1, acquiring an interactive image, and preprocessing the interactive image to obtain a first reference image;

[0067] S2, obtaining user needs, and performing feature extraction on the first reference image according to the user needs to obtain associated features and non-associated features;

[0068] S3, optimizing the first reference image according to the associated features and the non-associated features, and calibrating the optimized first reference image as the second reference image;

[0069] S4, showing the second reference image to the user, collecting user feedback and new requirements in real time, and iteratively optimizing the second reference image according to the user feedback and new requirements until the user no longer provides feedback or new requirements;

[0070] S5. Summarize the newly added requirements and user requirements into independent requirements, calibrate the second reference images iterated under the newly added requirements and user requirements into paired images corresponding to the independent requirements, and summarize the independent requirements and paired images into a reference image database.

[0071] As described in the above steps S1-S5, image processing is an increasingly developed technology that can provide people with a more convenient and efficient human-computer interaction experience. In this embodiment, it is first necessary to obtain images during the interaction process, which can be achieved through various image acquisition devices, such as cameras, scanners, etc. The acquired images may have problems such as noise and uneven lighting, so the images need to be preprocessed to improve the image quality. The preprocessing methods include denoising, smoothing, histogram equalization, etc., and finally a first reference image is obtained. After obtaining the first reference image, it is necessary to understand the user's needs, and based on the user's needs, feature extraction is performed on the first reference image, and the image is divided into associated features and non-associated features. Associated features refer to image features related to user needs, while non-associated features refer to image features that have little relationship with user needs. By extracting associated features and non-associated features, it is helpful to better understand user needs, thereby optimizing the image processing. According to the extracted associated features and non-associated features, the first reference image is optimized. The optimization methods may include image segmentation, feature fusion, image synthesis, etc. to meet the user's requirements for image quality, visual effects, etc. The optimized first reference image is calibrated as the second reference image and displayed to the user. In this process, user feedback and new requirements are collected in real time. According to the user feedback, the second reference image is iteratively optimized until the user no longer issues feedback and new requirements. This process reflects the real-time and dynamic nature of human-computer interaction, which helps to continuously improve image quality and user satisfaction. User feedback and new requirements are summarized to form independent requirements. The second reference image is corresponded to the independent requirements and calibrated as a paired image. Based on the independent requirements and paired images, a reference image database is constructed. The images in the database can provide reference for other users, thereby improving the efficiency and satisfaction of subsequent human-computer interaction.

[0072] In a preferred embodiment, the step of acquiring the interactive image and preprocessing the interactive image to obtain the first reference image includes:

[0073] S101, acquiring an interactive image, and performing denoising and enhancement processing on the interactive image;

[0074] S102, resizing and color correcting the interactive image after denoising and enhancement processing to obtain a front image;

[0075] S103 , performing partition processing on the front image, classifying the front image into a plurality of block images, and calibrating them as first reference images.

[0076] As described in the above steps S101-S103, in order to make better use of the interactive image, it is necessary to pre-process it to obtain a first reference image. First, various sensors and cameras are used to capture images of real-time scenes. These images usually contain rich scene information, but may also be affected by noise, lighting changes, etc. Therefore, the interactive image needs to be further denoised and enhanced. Denoising is to eliminate noise in the image and improve image quality. Enhancement is to enhance the image's contrast, brightness and other features to make the image easier to be recognized by the computer, which can effectively improve the usability of the interactive image. Then, the interactive image after denoising and enhancement is resized and color corrected. The size The purpose of adjustment is to make the image adapt to the needs of subsequent processing, such as image recognition and analysis, and color correction is to make the color of the image more realistic and full and improve the visual effect of the image. These two steps can make the interactive image more standardized. Then, the image after size adjustment and color correction is partitioned. Partitioning is to divide the image into multiple blocks so that different areas can be processed differently. In this process, the image can be divided into representative blocks according to the characteristics and needs of the image, which helps to extract key information from the image. Then, the partitioned image is calibrated as the first reference image to ensure a unified and stable reference in the subsequent processing process, thereby improving the processing effect and accuracy.

[0077] In a preferred embodiment, the step of performing partition processing on the front image and classifying the front image into a plurality of block images includes:

[0078] Get the size and color information of the front image;

[0079] According to the size information, the front image is divided into several regions of the same size and marked as initial region images;

[0080] According to the color information, color clustering analysis is performed on the initial region images to obtain the color distribution characteristics of each initial region image;

[0081] According to the color distribution characteristics, initial region images with similar color distribution in adjacent initial region images are merged to obtain multiple block images.

[0082] In this embodiment, after the front image is output, the size information and color information of the front image are obtained. The size information includes the aspect ratio and pixel resolution of the image. These parameters determine the basic form and detail expression of the image. The color information includes the color space, brightness, saturation, etc. of the image, which affects the visual effect and content expression of the image. According to the size information, the front image is divided into several areas of the same size and calibrated as initial area images. The purpose is to divide the image into small areas with the same characteristics for more in-depth analysis and processing. It should be noted that the size and number of the partitions should be adjusted according to actual needs and image characteristics. Then, according to the color information, the initial area image is subjected to color clustering analysis. Color clustering analysis is an unsupervised learning method that can classify similar colors into one category, so as to obtain the color distribution characteristics of each initial area image. Finally, according to the color distribution characteristics, the initial area images with similar color distribution in adjacent initial area images are merged to obtain multiple block images.

[0083] In a preferred embodiment, the step of performing color cluster analysis on the initial region images to obtain color distribution characteristics of each initial region image includes:

[0084] Get the color value of each pixel in the initial area image;

[0085] The number of clusters is obtained and marked as the benchmark parameter, and the color value consistent with the number of the benchmark parameter is randomly selected as the cluster center;

[0086] Count the frequency of each color cluster center appearing in the initial region image and mark it as the first condition parameter;

[0087] Obtaining a screening threshold, and comparing the screening threshold with the first condition parameter;

[0088] If the first condition parameter is greater than the screening threshold, the color corresponding to it is marked as the main color, and the color distribution characteristics of the initial area image are generated according to the main color;

[0089] If the first condition parameter is less than or equal to the screening threshold, the corresponding color is marked as a secondary color and is not recorded in the color distribution feature.

[0090] In the above, when performing color clustering analysis on the initial area image, it is first necessary to obtain the color value of each pixel from the initial area image. The color value is usually expressed in color space such as RGB (red, green, blue) or HSV (hue, saturation, brightness), and then the number of clusters needs to be determined. The number of clusters can be set according to the color distribution characteristics of the image, such as the commonly used contour coefficient, variance and other indicators to measure the clustering effect. After determining the number of clusters, it can be calibrated as a benchmark parameter, and then it is necessary to randomly select color values ​​consistent with the number of benchmark parameters as cluster centers. These cluster centers will serve as reference standards for color classification, and then count the frequency of each color cluster center in the initial area image and calibrate it as the first condition parameter, and then set the screening threshold. The screening threshold can be set according to the characteristics of the image. For example, it can be determined by a histogram or other method. By comparing the screening threshold with the first condition parameter, the primary color and the secondary color can be distinguished. If the first condition parameter is greater than the screening threshold, the corresponding color is marked as the primary color. The primary color is the color that is more concentrated in the image and has an important influence on the overall visual effect of the image. Based on the primary color, the color distribution characteristics of the initial area image can be generated. On the contrary, if the first condition parameter is less than or equal to the screening threshold, the corresponding color is marked as the secondary color. The distribution of the secondary color in the image is relatively dispersed and has less influence on the overall visual effect of the image. When generating the color distribution characteristics, these secondary colors can be ignored.

[0091] In a preferred embodiment, the step of merging initial region images with similar color distribution in adjacent initial region images according to the color distribution characteristics includes:

[0092] Acquire the color distribution characteristics of the adjacent initial region images, and synchronously perform vectorization processing to obtain the second conditional parameter;

[0093] Obtain a standard function, input the second condition parameter into the standard function, and mark its output result as an evaluation parameter;

[0094] Obtaining an evaluation threshold, and comparing the evaluation threshold with an evaluation parameter;

[0095] If the evaluation parameter is greater than the evaluation threshold, the corresponding adjacent initial area images are merged to obtain a merged block image;

[0096] If the evaluation parameter is less than or equal to the evaluation threshold, a splicing curve is constructed at the connection of the corresponding adjacent initial region images.

[0097] In this embodiment, after the color distribution features are determined, the color distribution features of the adjacent initial region images are subjected to synchronous vectorization processing to obtain a second condition parameter, and then a standard function is obtained. The expression of the standard function is: , where represents the evaluation parameters, represents the number of second conditional parameters, and Represents the color distribution characteristics of adjacent initial region images, inputs the second condition parameter into the standard function, and obtains the evaluation parameter. The evaluation parameter can be used to measure the similarity of adjacent initial region images. The higher the value, the more similar the adjacent initial region images are. Then, the evaluation parameter is compared with the evaluation threshold through a preset evaluation threshold. The selection of the evaluation threshold can be adjusted according to the actual application scenario and needs. If the evaluation parameter is greater than the evaluation threshold, it means that the adjacent initial region images have a high similarity and can be merged. After the adjacent initial region images are merged, a merged block image is obtained. If the evaluation parameter is less than or equal to the evaluation threshold, it means that the adjacent initial region images have a low similarity and cannot be directly merged. In this case, it is necessary to construct a splicing curve at the connection of the corresponding adjacent initial region images. The splicing curve can smoothly connect the edges of adjacent images for easy distinction.

[0098] In a preferred embodiment, the step of obtaining user requirements and extracting features from the first reference image according to the user requirements to obtain associated features and non-associated features includes:

[0099] S201, collecting voice information or text information corresponding to user needs, wherein the voice information is synchronously converted into text information;

[0100] S202, extracting a block image corresponding to the user's needs from the first reference image, and performing an offset process on the edge curve of the block image to obtain a feature extraction curve;

[0101] S203 , performing image segmentation on the first reference image according to the feature extraction curve, and marking the block image within the feature extraction curve as the associated feature, and marking the block image outside the feature extraction curve as the non-associated feature.

[0102] As described in the above steps S201-S203, when collecting user needs, it is necessary to collect voice information or text information corresponding to the user needs. In this process, the voice information will be synchronously converted into text information, and then the block image corresponding to the user needs is extracted from the first reference image to find the area related to the user needs. After that, the edge curve of the block image is offset to obtain the feature extraction curve after the edge curve offset, and then the first reference image is performed image segmentation according to this curve. Image segmentation is the process of dividing an image into multiple areas, which helps to better analyze various features in the image. In this process, the block image within the feature extraction curve is calibrated as associated features, and these associated features are extracted separately, so that the second reference image can be obtained. These associated features are directly related to the user needs, while the block images outside the feature extraction curve are calibrated as non-associated features, which have no direct relationship with the user needs.

[0103] In a preferred embodiment, the step of iteratively optimizing the second reference image according to user feedback and new requirements until the user no longer provides feedback or new requirements includes:

[0104] S401, displaying the second reference image to the user, and collecting the user's feedback and new requirements in real time;

[0105] S401: If the user does not send any feedback or additional requirements, the second reference image is recorded as a standard image and added to the reference image database;

[0106] S403: If the user issues feedback and new requirements, the second reference image is synchronously optimized according to the user's feedback and new requirements, including adjusting image parameters, size, and content, and outputting the optimized image;

[0107] S404, collecting user feedback and new requirements under the optimized image, and stopping the optimization when the user no longer provides feedback and new requirements. Otherwise, continue to output the optimized image until the user no longer provides feedback and new requirements, and record the optimized image under each new requirement and add it to the benchmark image database.

[0108] As described in the above steps S401-S404, during the image processing process, the second benchmark image needs to be continuously improved according to the user's feedback and new requirements. First, the second benchmark image is displayed to the user so that the user can have an intuitive understanding, and the user's feedback and new requirements are collected in real time. If the user does not provide feedback and new requirements after observing and evaluating the second benchmark image, then the image can be recorded as a standard image and included in the benchmark image database, which means that the second benchmark image has met the user's basic needs and can be used as a reference standard for other similar images. However, if the user provides feedback and new requirements on the second benchmark image, then the image needs to be improved according to these opinions and requirements. Like synchronous optimization, this optimization process includes adjusting the parameters, size and content of the image to ensure that the image better meets the needs of users. After the optimization is completed, a new optimized image will be output. After the optimized image is output, it is necessary to collect user feedback and new requirements again. This process will continue until the user no longer provides feedback and new requirements, which means that the user is satisfied with the optimized image. At this time, the optimization can be paused. During each iterative optimization process, the user's new requirements and the corresponding optimized images will be recorded. These optimized images can be used as a reference for future iterative optimization, which will help improve the quality of the benchmark image. At the same time, incorporating these optimized images into the benchmark image database will also help optimize other similar images.

[0109] In a preferred embodiment, the step of aggregating the reference image database according to the independent requirements and paired images includes:

[0110] S501, classify and sort out the newly added requirements and user requirements, group the same or similar requirements into one category, and obtain an independent requirement list;

[0111] S502, matching each independent requirement in the independent requirement list with the second reference image, and finding a paired image that best matches the independent requirement;

[0112] S503: Associating the paired images with the corresponding independent requirements and aggregating them into a reference image database.

[0113] As described in the above steps S501-S50, constructing a complete benchmark image database can provide corresponding reference basis for subsequent image processing. For example, when different users have the same requirements for the same image, the corresponding optimized image can be called from the benchmark image database for display, thereby improving the efficiency of human-computer interaction. In the process of human-computer interaction, user requirements are continuously collected, and the collected requirements are classified and sorted, and the same or similar requirements are classified into one category. After the sorting is completed, an independent requirement list will be obtained, which contains independent and non-repetitive requirements, and the corresponding second benchmark images are matched one by one. Then, a benchmark image database can be established, so that when the requirements in the independent requirement list are subsequently accepted, the corresponding second benchmark image can be quickly matched and displayed to the user, thereby improving the efficiency and accuracy of human-computer interaction.

[0114] See also Figure 2 , a human-computer interaction optimization system based on image processing, applied to the above-mentioned human-computer interaction optimization method based on image processing, comprising:

[0115] An image acquisition module, the image acquisition module is used to acquire an interactive image and pre-process the interactive image to obtain a first reference image;

[0116] A feature extraction module, the feature extraction module is used to obtain user requirements, and extract features from the first reference image according to the user requirements to obtain associated features and non-associated features;

[0117] An optimization module, the optimization module is used to optimize the first reference image according to the associated features and the non-associated features, and calibrate the first reference image after the optimization as the second reference image;

[0118] An iteration module, the iteration module is used to display the second reference image to the user, collect user feedback and new requirements in real time, and iteratively optimize the second reference image according to the user feedback and new requirements until the user no longer sends feedback and new requirements;

[0119] The database construction module is used to aggregate the new requirements and user requirements into independent requirements, calibrate the second reference images iterated under the new requirements and user requirements into paired images corresponding to the independent requirements, and aggregate the independent requirements and paired images into a reference image database.

[0120] In the above, the system includes an image acquisition module, a feature extraction module, an optimization module, an iteration module and a database construction module. The image acquisition module is responsible for acquiring image information during the interaction process. Through a high-definition image acquisition device, the interaction scene between the user and the device can be captured. In order to ensure the image quality, it also has a pre-processing function to perform denoising, enhancement and other processing on the collected image to make the image clearer, which is conducive to subsequent feature extraction and optimization processing. The feature extraction module is responsible for extracting features related to user needs from the pre-processed image. These features may include the texture, color, shape, etc. of the image. By analyzing these features, the user's needs can be better understood. At the same time, the feature extraction module will also distinguish between related features and non-related features for subsequent optimization processing. The optimization module is used to optimize the first reference image according to the associated features and the non-associated features to obtain a second reference image that better meets the needs of the user. The iteration module is responsible for collecting the user's feedback and new requirements in real time. According to the user's feedback and new requirements, the iteration module will iteratively optimize the second reference image until the user no longer issues feedback and new requirements. The database construction module is responsible for aggregating the new requirements and user requirements into independent requirements, and calibrating the iterated second reference image into a paired image corresponding to the independent requirements. According to the independent requirements and paired images, the database construction module will also summarize them into a reference image database to provide support for subsequent human-computer interaction, which can not only improve the interaction efficiency, but also provide users with a more personalized interaction experience.

[0121] See also Figure 3 , a human-computer interaction optimization device based on image processing, comprising:

[0122] at least one processor;

[0123] and a memory communicatively coupled to the at least one processor;

[0124] The memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the above-mentioned human-computer interaction optimization method based on image processing.

[0125] In the above, the processor may include a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), etc., which are responsible for executing instructions in the computer program to realize the various functions of the microwave ablation system. The memory may include a read-only memory (ROM), a random access memory (RAM), a flash memory, etc., which are used to store computer programs and various data. The computer programs in the memory may include an operating system, an application program, etc., which are executed by the processor to realize the various functions of the microwave ablation system. In addition, the electronic device may also include an input device, an output device, a network interface, etc. The input device may include a storage medium, a drive, etc., for receiving user input, the output device may include a display, an audio device, etc., for displaying information and output results to the user, and the network interface may include a wired network interface and a wireless network interface.

[0126] 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 includes 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 presence of other identical elements in the process, device, article or method including the element.

[0127] The above is only a preferred embodiment of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be considered as the protection scope of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention shall be implemented according to the conventional means in the art unless otherwise specified and limited.

Claims

1. A human-computer interaction optimization method based on image processing, Features: include: Acquire an interactive image, and preprocess the interactive image to obtain a first reference image; Obtaining user needs, and performing feature extraction on the first reference image according to the user needs to obtain associated features and non-associated features; Optimizing the first reference image according to the associated features and the non-associated features, and calibrating the optimized first reference image as the second reference image; The second reference image is displayed to the user, and user feedback and new requirements are collected in real time, and the second reference image is iteratively optimized according to the user feedback and new requirements until the user no longer provides feedback and new requirements; Aggregating the newly added requirements and user requirements into independent requirements, calibrating the iterated second reference images under the newly added requirements and user requirements as paired images corresponding to the independent requirements, and aggregating the independent requirements and paired images into a reference image database; The step of acquiring the interactive image and preprocessing the interactive image to obtain the first reference image includes: Acquire the interactive image, and perform denoising and enhancement processing on the interactive image; Resize and color correct the interaction image after denoising and enhancement to obtain a front image; Performing partition processing on the front image, classifying the front image into a plurality of block images, and marking them as first reference images; The step of performing partition processing on the front image and classifying the front image into a plurality of block images comprises: Acquire size information and color information of the front image; According to the size information, the front image is equally divided into a number of regions of the same size and marked as initial region images; Performing color cluster analysis on the initial region images according to the color information to obtain color distribution characteristics of each initial region image; According to the color distribution feature, initial region images with similar color distribution in adjacent initial region images are merged to obtain a plurality of block images; The step of performing color cluster analysis on the initial region images to obtain color distribution characteristics of each initial region image comprises: Obtaining the color value of each pixel in the initial area image; The number of clusters is obtained and marked as a reference parameter, and a color value consistent with the number of the reference parameter is randomly selected as a cluster center; Counting the frequency of each color cluster center appearing in the initial region image and marking it as the first condition parameter; Obtaining a screening threshold, and comparing the screening threshold with a first condition parameter; If the first condition parameter is greater than the screening threshold, the color corresponding to the first condition parameter is marked as the main color, and the color distribution characteristics of the initial area image are generated according to the main color; If the first condition parameter is less than or equal to the screening threshold, the corresponding color is marked as a secondary color and is not recorded in the color distribution feature; The step of merging the initial region images with similar color distribution in adjacent initial region images according to the color distribution characteristics comprises: Acquire the color distribution characteristics of the adjacent initial region images, and synchronously perform vectorization processing to obtain the second conditional parameter; Obtaining a standard function, and inputting the second condition parameter into the standard function, and marking the output result as an evaluation parameter; Obtaining an evaluation threshold, and comparing the evaluation threshold with an evaluation parameter; If the evaluation parameter is greater than the evaluation threshold, the corresponding adjacent initial area images are merged to obtain a merged block image; If the evaluation parameter is less than or equal to the evaluation threshold, constructing a splicing curve at the connection of the corresponding adjacent initial area images; The step of obtaining user requirements and extracting features from the first reference image according to the user requirements to obtain associated features and non-associated features includes: Collecting voice information or text information corresponding to the user's needs, wherein the voice information will be synchronously converted into text information; Extracting a block image corresponding to the user's needs from the first reference image, and performing an offset process on the edge curve of the block image to obtain a feature extraction curve; Performing image segmentation on the first reference image according to the feature extraction curve, and marking the block image within the feature extraction curve as the associated feature, and marking the block image outside the feature extraction curve as the non-associated feature; The step of iteratively optimizing the second reference image according to the user feedback and the new requirements until the user no longer provides feedback and new requirements includes: Displaying the second reference image to the user, and collecting the user's feedback and new requirements in real time; If the user does not send any feedback or additional requirements, the second reference image is recorded as the standard image and added to the reference image database; If the user issues feedback and new requirements, synchronously optimizing the second reference image according to the user's feedback and new requirements, including adjusting image parameters, size and content, and outputting the optimized image; Collecting user feedback and new requirements under the optimized image, and stopping the optimization when the user no longer issues feedback and new requirements, otherwise, continue to output the optimized image until the user no longer issues feedback and new requirements, and record the optimized image under each new requirement and add it to the reference image database; The step of aggregating the independent requirements and paired images into a reference image database comprises: Classify and sort out the newly added requirements and user requirements, group the same or similar requirements into one category, and obtain an independent requirement list; Matching each independent requirement in the independent requirement list with the second reference image to find a paired image that best matches the independent requirement; The paired images are associated with corresponding independent requirements and summarized into a reference image database.

2. A human-computer interaction optimization system based on image processing, applied to the human-computer interaction optimization method based on image processing described in claim 1, Features: include: An image acquisition module, the image acquisition module is used to acquire an interactive image and preprocess the interactive image to obtain a first reference image; A feature extraction module, the feature extraction module is used to obtain user requirements, and extract features from the first reference image according to the user requirements to obtain associated features and non-associated features; An optimization module, the optimization module is used to optimize the first reference image according to the associated features and the non-associated features, and calibrate the first reference image after the optimization as the second reference image; an iteration module, the iteration module being used to display the second reference image to the user, collect user feedback and new requirements in real time, and iteratively optimize the second reference image according to the user feedback and new requirements until the user no longer issues feedback and new requirements; A database construction module, the database construction module is used to aggregate the new requirements and user requirements into independent requirements, calibrate the second reference images iterated under the new requirements and user requirements as paired images corresponding to the independent requirements, and aggregate the independent requirements and paired images into a reference image database.

3. A human-computer interaction optimization device based on image processing, Features: include: at least one processor; and a memory communicatively coupled to the at least one processor; Wherein, the memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the human-computer interaction optimization method based on image processing described in claim 1.