An Image Retrieval Method Based on Multi-feature Fusion
A multi-feature fusion and image retrieval technology, applied in digital data information retrieval, special data processing applications, instruments, etc., can solve problems such as imperfect retrieval methods, improve actual search experience, improve search efficiency, and improve single feature coverage Insufficient effect
Active Publication Date: 2021-06-01
上海得见计算机科技有限公司
View PDF8 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0006] The present invention proposes an image retrieval method based on multi-feature fusion, which solves the problem of incomplete fusion of multiple image feature retrieval methods in the prior art
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0048]An image retrieval method based on multi-feature fusion, the specific steps are as follows:
[0049](1) Get the target image I, and calculate the image characteristics of the target image i
[0050]1) Extract the color characteristics of the target image i, and store the color characterization of the image
[0051]1 Convert the target image i from the RGB color space to the HSV color space according to the standard conversion formula;
[0052]2 Target image I convert to the HSV color space according to the formulawithThe first step is quantified, and the hue H is quantified to seven intervals, the brightness V and the saturation S are quantified to 3 intervals;
[0053]3 According to the formula L = 9h + 3s + V, the RGB value of the target image I is mapped to the 63 color space of the HSV;
[0054]The amount of information supplied in different positions in a piece of image is different. The information of usually a pair image is mainly concentrated in the center of the image, and the edge por...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention proposes a kind of image retrieval method based on multi-feature fusion, and its steps are as follows: obtain target image I, and calculate the image feature of target image I: Extract the color feature of target image I, and store in the color feature storehouse of image; Extract the shape feature of the target image I and store it in the shape feature library of the image; extract the texture feature of the target image I and store it in the texture feature library of the image; calculate the similarity between the target image I and the image data set to obtain the final retrieval result . The beneficial effects of the present invention are as follows: the user can obtain similar home scheme images according to the retrieved images. This method aims at the deficiency of the existing single feature, and deals with the combination of feature sets in the home furnishing industry. In the actual home scene, it can improve the efficiency of single feature retrieval and improve the problem of insufficient single feature coverage. It can improve the search efficiency of home product images, save the time for users to search for home solutions, and improve the actual search experience of users.
Description
Technical field[0001]The present invention relates to the field of image retrieval techniques, particularly an image retrieval method based on a multi-feature fusion.Background technique[0002]The content-based image retrieval method refers to information such as visual features and spatial relationships such as the image itself, establishes a high-dimensional feature vector library of an image, and a user's image retrieval result is returned according to the high-dimensional feature vector of the image. The method is more effective than the text-based image retrieval method.[0003]Image features can be divided into color features, texture feature, and shape features of color characteristics, texture feature, and shape features based on logical types. The color characteristics are one of the most widely used image features, and can provide image search for image-based functions; texture features focus on the texture mode in the image block; the shape features mainly describe the struc...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/583G06K9/46G06K9/62
CPCG06V10/56G06F18/253
Inventor 栾雄张闻强徐念龙杨莹
Owner 上海得见计算机科技有限公司