Scene image classifying method based on annular space pyramid and multi-kernel study

Abstract

The invention discloses a scene image classifying method based on an annular space pyramid and multi-kernel study. The scene image classifying method comprises the following steps: establishing a training image set and a testing image set; carrying out a multi-feature extraction stage: including extracting a Dense-SIFT feature, an L-Gist feature and a colored feature; training a dictionary by virtue of K-means++ clustering, carrying out secondary clustering on each extracted feature, and carrying out secondary clustering on a vision dictionary set generated in first clustering, so as to obtain a total vision dictionary; carrying out an image feature coding stage: carrying out annular space pyramid division on the image, forming a vector expression form for each sub-image block divided by the pyramid based on the vision dictionary; carrying out a multi-kernel study stage: diving the image by virtue of the annular space pyramid, and respectively distributing a kernel function for each sub-image block and each colored feature; and carrying out a classification judging stage. Compared with a conventional single feature method, the scene image classifying method has the advantages that a scene image is represented by a complementary combination of the Dense-SIFT feature, the L-Gist feature and an HSV global color feature, so that the complete information of the image can be effectively represented, and the scene classification can be well realized.

Description

technical field [0001] The invention belongs to the field of machine learning and digital image processing, in particular to a method for classifying scene images based on annular space pyramid and multi-core learning Background technique [0002] In recent years, due to the rapid development of multimedia and Internet technology, the rapid expansion of image information resources has been greatly promoted. While massive image resources bring great convenience to our work and life, how to manage and quickly retrieve our Images of interest are becoming more and more difficult. Therefore, in the face of vast image resources, relying on the traditional manual labeling method is not only time-consuming and laborious, but also has subjective uncertainty, which obviously does not meet the needs of the rapid development of today's multimedia information age. Then, how to use smart devices such as computers to complete the automatic classification and efficient management of image ...

AI Technical Summary

Problems solved by technology

[0005] In view of the above technical problems, the present invention proposes a scene image classification method based on annular space pyramid and multi-core learning, extracts local feature Dense-SIFT and local Gist feature L-Gist from the scene image, and combines the global color feature of HSV color space To combine and represent scene images, it overcomes the problem of information loss caused by using a single feature to represent images in traditional classification methods; uses the encoding method of three-level spatial pyramid aggregation to encode these features; in order to increase the spatial information when classifying scene images and each image Blocks have different contributions during classification, and the method of dividing and weighting the ring space pyramid is used to increase the spatial information between the scene image features; Each image patch of is assigned a kernel function, and by learning the weights of each kernel, the synthetic kernel with the strongest distinguishing ability is obtained

Images