Colorful animal image retrieval method based on content and colorful animal image retrieval system based on content

Abstract

The invention provides a colorful animal image retrieval method based on content and a colorful animal image retrieval system based on content. The method comprises the following steps: step 1, inputting an inquiry image which serves as a retrieval object; step 2, extracting the color, grain and framework characteristic descriptions of interested content in the inquiry image based on the guidance of a characteristic template library, and performing normalization processing, wherein the characteristic descriptions, which are different in animal color, grain and framework and are formed on the basis of sample statistics, are stored in the characteristic template library; step 3, performing retrieval based on an R-tree index structure on the characteristic descriptions extracted in the step 2, performing search calculation on image characteristic indexes in a characteristic index library, and extracting a corresponding image from an image database according to the matching result. The method and the system can be used for overcoming the influence of illumination changes on image retrieval judgment to a certain extent, describing the angle and posture semantic information of an inquiry animal according to the framework characteristic analysis during retrieval, and generating a retrieval result consistent with the human visual perception.

Description

technical field [0001] The invention belongs to the technical field of content-based image retrieval, in particular to a content-based color animal image retrieval method and system. Background technique [0002] Content-based image retrieval (CBIR) is an image retrieval technology that uses the visual features of images (color, texture, shape, spatial location, etc.) to search for images. It combines artificial intelligence, object-oriented technology, cognitive psychology, database and other multidisciplinary knowledge, and automatically extracts the underlying visual features and high-level semantic features of images, thus avoiding a series of problems generated by image retrieval systems based on manual annotation. The problem makes the application of large-scale image database system more realistic. Therefore, with the rapid growth of various image resources in the Internet age, CBIR has become the mainstream technology in the field of image retrieval research in the ...

AI Technical Summary

Problems solved by technology

[0009] 1. Most of the existing CBIR technologies based on the underlying features of images such as color, texture, and inflection point features are extremely sensitive to image scale and brightness changes

On the other hand, the animal images in the natural background can easily change the above-mentioned underlying characteristics of the animal body surface due to reasons such as illumination and occlusion.

Therefore, the CBIR technology that uses the underlying features of images such as color, texture, and inflection point features will cause a high probability of misjudgment when retrieving color animal images;

[0010] 2. Most of the existing CBIR technologies based on visual features containing certain semantic information such as object shape and spatial position are mainly aimed at grayscale images, and there is relatively little work on segmenting color images.

On the other hand, this type of retrieval technology mainly relies on the quality of image segmentation, and the existing image segmentation methods are prone to over-segmentation problems for animal image segmentation under changing illumination conditions, which greatly reduces the precision rate;

[0011] 3. CBIR based on the high-level semantics of images is still in the research stage. Although the academic community is working hard to study the high-level semantic expression of images, there is still no general recognition of what the high-level semantics of images are and how to express these semantics. Mathematical or perceptual models, there is still a certain distance from the application;

[0012] 4. When most of the existing CBIR technologies are applied to the retrieval of color animal images, they lack the characteristic description of animal posture, and the retrieval results cannot well meet the retrieval requirements.

Images