No abstract

Although current gymnastics action detection algorithm has good detection and recognition results but cannot effectively identify a variety of consecutive gymnastic actions and many gymnastics has high false rate. So on this paper we improve the CRF model and bag-of-visual-words semantic model, combine the advantages of both models to build a hierarchical model for behavior recognition, first we create a hierarchical semantic mark CRFs model, the model is divided into upper and lower layers and a gymnastic image filter that based on bag-of-visual-words semantic model. Identifying the error action image by the semantic, not only in line with the cognitive process of machine vision, and can effectively compensate the existing algorithm correcting the high false positives rate. Experiments show that by combining the two algorithms we can detect errors gymnastics image effectively, recognition rate compared with other algorithms improved, and the false detection rate reduced.

This paper aims to overcome two major defects with the traditional rock image classification framework based on convolutional neural network (CNN), namely, slow training and poor classification accuracy. First, the causes of the two defects were analyzed in details. Through the analysis, the slow training is attributable to the information redundancy in the original image, and the classification error to the lack of differentiation of rock features extracted from the spatial domain. Therefore, the original image was divided into multiple blocks of equal size, and the discrete cosine transform (DCT) was introduced to process each block. After the transform, ten or fifteen frequency coefficients in the upper left corner of the 2D frequency coefficient matrix were retained, and added to the traditional CNN framework for image classification. Experimental results show that the proposed DCT-CNN framework outperformed the CNN framework in training time and classification accuracy.

No abstract