Picture generation method based on image analysis technology
Through the method based on image analysis technology, through image data acquisition, feature extraction and text information fusion, the problem of inconsistency between image rendering and artistic information in the prior art is solved, and a consistent artistic image is generated.
Patent Information
- Application Number
- CN202411259121.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-10
- Publication Date
- 2024-12-06
AI Technical Summary
In the existing security monitoring image analysis technology, although the generated images have consistent characteristics, the rendering and artistic information of the paintings are difficult to be consistent, and cannot meet artistic needs.
Using a method based on image analysis technology, similar images are generated through image data acquisition, feature extraction, similarity calculation, text information extraction and fusion drawing to ensure that the rendering and artistic information of the painting are consistent. Specific steps include image data acquisition, feature map drawing and comparison, single image locking, text information extraction and similar image drawing, and the optimization and organization of image parameters using multi-dimensional vectorization and convolution kernel calculations.
It realizes the desired image drawing as needed, ensuring that the rendering and artistic information of the generated image are consistent and meet artistic needs.
Smart Images

Figure CN119091005A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of picture generation, and in particular to a picture generation method based on image analysis technology. Background Art
[0002] Image analysis technology, also known as scene analysis or image understanding, is a key technology in the field of computer vision. It mainly uses mathematical models and combines image processing technology to analyze the underlying features and upper structure of images, thereby extracting information with certain intelligence;
[0003] In the existing security monitoring image analysis application technology, picture drawing is an extremely important component. The commonly used method is to extract features from existing pictures and find similar images in the database. However, in actual artistic needs, the generated images are only consistent with the feature images, but the rendering of the paintings and the artistic information to be expressed are difficult to be consistent. Therefore, in order to solve the above problems, a picture generation method based on image analysis technology is very important. Summary of the invention
[0004] 1. Technical issues to be resolved
[0005] In view of the deficiencies in the prior art, the present invention provides a method for generating a picture based on image analysis technology, which solves the problems existing in the above-mentioned background technology.
[0006] (II) Technical solution
[0007] To achieve the above object, the present invention provides the following technical solution: a method for generating a picture based on image analysis technology, comprising the following steps:
[0008] S1. Image data acquisition: Use a camera to take a picture of the painting, and use a feature extractor to lock a single image feature and extract the image feature;
[0009] S2, feature map drawing and comparison: generate a single image by fusing the extracted image feature maps, and compare and analyze the image with the feature maps stored in the database to calculate the similarity;
[0010] S3, single image locking: when the similarity reaches the preset value, the image feature information is determined, and the single image is locked, the color and brightness information of the image is further extracted, and the data is integrated;
[0011] S4, text information extraction: extract and analyze the text information to find the corresponding image feature data;
[0012] S5. Similar picture drawing: The integrated data is merged with the data obtained from the text information, and similar images are drawn based on the image information calculated from the photographed paintings.
[0013] Furthermore, the feature extractor in S1 receives the light wave signal reflected by the painting, obtains the corresponding color blocks according to the frequency, and calculates the feature information according to the distribution of the color blocks.
[0014] Furthermore, in S2, the extracted image is vectorized in multiple dimensions, and then the image is compared with the multidimensional vector data information of the image features stored in the database to obtain a similarity value d. The specific calculation method is:
[0015]
[0016] Among them, x i and i are the i-th components of the two eigenvectors, n is the dimension of the eigenvector, the smaller the value of d is, the more similar the corresponding features are, and the vector values of multiple features are substituted to integrate the obtained multiple d values.
[0017] Furthermore, in S3, the average value of the similarity value d is compared and analyzed with the preset value D. When the value of d is less than D, it is locked as the same feature data, stored, and displayed as feature text.
[0018] Furthermore, in S4, segmentation, integration and information extraction are performed according to the provided text information to find the corresponding feature image in the database.
[0019] Furthermore, the specific function calculation formula in S4 is:
[0020]
[0021] Among them, O H and O W is the height and brightness of the output image, I H and I W is the height and brightness of the input image, K H and K W is the height and brightness of the convolution kernel, P H and P W is the filling size, S is the stride, the fused feature image value is input into the function formula, and compared with the captured image to calculate P H and P W , and then optimize and organize the pictures so that the basic parameters of the two pictures are consistent.
[0022] Furthermore, by photographing the painting, the information of a single feature image is extracted, and the text information is segmented, integrated and extracted to form feature text. The corresponding feature information is found in the database, and finally the brightness of the painting and other information are analyzed and integrated with the feature information to draw a similar image, effectively realizing the drawing of the desired picture according to the needs.
[0023] (III) Beneficial effects
[0024] The present invention provides a method for generating pictures based on image analysis technology, which has the following beneficial effects:
[0025] By photographing the paintings, the information of a single feature image is extracted, and the text information is segmented, integrated and extracted to form feature text. The corresponding feature information is found in the database, and finally the brightness of the painting and other information are analyzed and integrated with the feature information to draw a similar image, effectively realizing the drawing of the desired picture according to the needs. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] Figure 1 The present invention is a flow chart of a method for generating pictures based on image analysis technology. DETAILED DESCRIPTION
[0027] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. 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.
[0028] The present invention provides a technical solution: a method for generating a picture based on image analysis technology, comprising the following steps:
[0029] S1. Image data acquisition: Use a camera to photograph the painting, and use a feature extractor to lock a single image feature and extract the image feature. The feature extractor receives the light wave signal reflected by the painting, obtains the corresponding color block according to the frequency, and calculates the feature information based on the distribution of the color block.
[0030] S2. Feature map drawing and comparison: A single image is generated by fusing the extracted image feature maps, and the image is compared and analyzed with the feature maps stored in the database to calculate the similarity. The extracted image is vectorized in multiple dimensions, and the similarity value d is calculated and compared with the image feature multidimensional vector data information stored in the database. The specific calculation method is:
[0031]
[0032] Among them, x i and i are the i-th components of the two eigenvectors, n is the dimension of the eigenvector, the smaller the value of d is, the more similar the corresponding features are, and the vector values of multiple features are substituted to integrate the obtained multiple d values.
[0033] S3. Single image locking: When the similarity reaches the preset value, the image feature information is determined and the single image is locked. The color and brightness information of the image is further extracted and the data is integrated. The average value of the similarity value d is compared with the preset value D. When the value of d is less than D, it is locked as the same feature data and stored to form feature text for display.
[0034] S4. Text information extraction: Extract and analyze the text information to find the corresponding image feature data, perform segmentation, integration and information extraction based on the provided text information, and find the corresponding feature image in the database. The specific function calculation formula is:
[0035]
[0036] Among them, O H and O W is the height and brightness of the output image, I H and I W is the height and brightness of the input image, K H and K W is the height and brightness of the convolution kernel, P H and P W is the filling size, S is the stride, the fused feature image value is input into the function formula, and compared with the captured image to calculate P H and P W , and then optimize and organize the pictures so that the basic parameters of the two pictures are consistent.
[0037] S5. Drawing of similar pictures: The integrated data is merged with the data obtained from the text information, and similar images are drawn based on the image information calculated from the photographed paintings. The information of a single feature image is extracted by photographing the paintings, and the text information is segmented, integrated and extracted to form feature text. The corresponding feature information is found in the database, and finally the brightness of the painting and other information are analyzed and merged with the feature information to draw similar images, which effectively realizes drawing the desired pictures according to the needs.
[0038] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0039] 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. A method for generating a picture based on image analysis technology, characterized in that: The following steps are involved: S1. Image data acquisition: Use a camera to take a picture of the painting, and use a feature extractor to lock a single image feature and extract the image feature; S2, feature map drawing and comparison: generate a single image by fusing the extracted image feature maps, and compare and analyze the image with the feature maps stored in the database to calculate the similarity; S3, single image locking: when the similarity reaches the preset value, the image feature information is determined, and the single image is locked, the color and brightness information of the image is further extracted, and the data is integrated; S4, text information extraction: extract and analyze the text information to find the corresponding image feature data; S5. Similar picture drawing: Fuse the integrated data with the data obtained from the text information, and draw similar images based on the image information calculated from the photographed paintings.
2. The method for generating a picture based on image analysis technology according to claim 1, characterized in that: The feature extractor in S1 receives the light wave signal reflected by the painting, obtains the corresponding color blocks according to the frequency, and calculates the feature information according to the distribution of the color blocks.
3. The method for generating a picture based on image analysis technology according to claim 1, characterized in that: In S2, the extracted image is vectorized in multiple dimensions, and then the image is compared with the multidimensional vector data information of the image features stored in the database to obtain the similarity value d. The specific calculation method is: Among them, x i and i are the i-th components of the two eigenvectors, n is the dimension of the eigenvector, the smaller the value of d is, the more similar the corresponding features are, and the vector values of multiple features are substituted to integrate the multiple d values obtained.
4. The method for generating a picture based on image analysis technology according to claim 1, characterized in that: In the above S3, the average value of the similarity value d is compared and analyzed with the preset value D. When the value of d is smaller than D, it is locked as the same feature data, stored, and displayed as feature text.
5. The method for generating a picture based on image analysis technology according to claim 1, characterized in that: In S4, segmentation, integration and information extraction are performed according to the provided text information, and the corresponding feature image is found in the database.
6. The method for generating a picture based on image analysis technology according to claim 1, characterized in that: The specific function calculation formula in S4 is: Among them, O H and O W is the height and brightness of the output image, I H and I W is the height and brightness of the input image, K H and K W is the height and brightness of the convolution kernel, P H and P W is the filling size, S is the stride, the fused feature image value is input into the function formula, and compared with the captured image to calculate P H and P W , and then optimize and organize the pictures so that the basic parameters of the two pictures are consistent.
7. The method for generating a picture based on image analysis technology according to claim 6, characterized in that: By photographing the paintings, the information of a single feature image is extracted, and the text information is segmented, integrated and extracted to form feature text. The corresponding feature information is found in the database, and finally the brightness of the painting and other information are analyzed and integrated with the feature information to draw a similar image, effectively realizing the drawing of the desired picture according to the needs.
