Scene classification method based on Gist characteristics and extreme learning machine

Abstract

The invention provides a scene classification method based on Gist characteristics and an extreme learning machine. The scene classification method comprises the steps of firstly extracting the Gist characteristics of images, performing convolution operation on the scene images by use of Gabor wavelet, next, extracting Gist vectors as characteristic descriptions of the scene images and applying the characteristic descriptions to scene classification. According to the scene classification method, a comprehensive cognition is generated on images by use of the Gist characteristics, and five natural attributes, namely naturalness, opening degree, roughness, expanding degree and bumpiness, are comprehensively described, and compared with the traditional scene classification method, setting different parameters and threshold according to various practical situations can be avoided. The technical problem that the classification threshold must be adjusted continuously under the condition of many change conditions during traditional classification is solved; the scene classification method is high in operation speed, excellent in generalization ability, and excellent in expansibility, and shows better superiority with the increase of the classification situation complexity and the expansion of the scale.

Description

technical field [0001] The invention relates to a scene image feature extraction method and an extreme learning machine classification method. Specifically, it is to use the Gist global feature of the extracted image, and then use the extreme learning machine trained by the sample to classify the image. Background technique [0002] Scene classification is an important branch of computer vision image classification. The concept of scene description and understanding was further clarified at the 2006 MIT Scene Understanding Symposium. At the same time, it was also pointed out that scene classification is a new promising research direction, and its main applications are in four main areas: image / video retrieval, computer vision tasks, mobile robotics, and image enhancement. [0003] Vision-based scene classification methods can be roughly divided into three categories: object-based scene classification, region-based object classification, and context-based scene classifica...

AI Technical Summary

Problems solved by technology

Such methods are not so flexible in the face of segmentation, which is especially prominent in unconstrained settings, and therefore, there are few fully region-based scene classification methods

Images