Image Feature Extraction And Its Application In Cable Fault Signal Recognition

With the rapid development of multimedia technology, digital technology and the Internet applications, multimedia information have been widely applied in various fields, such as digital libraries, medical images, satellite remote sensing images, and the public security system. How to find the desired image rapidly and accurately from the vast ocean is an urgent problem need to solve.First, the three types of commonly used image retrieval image features, color features, shape features and texture features,were introduced. As the image retrieval based on single feature using only part of the image information, so the search results are not very satisfactory. With the deep application of image database the contradiction increasingly prominent. In order to improve the average recall rate and the average precision rate of image retrieval, an improved fusion algorithm of the weighted features is presented. Firstly, the shape features and the texture features of images are extracted. Then the elements of the vectors are normalized respectively. In the next step, the Euclidian distance and the squared Euclidian distance and the City-Block distance are calculated. The Mean values of the 3 kinds of distances are obtained and used as the shape similarity distance and the texture similarity distance. Finally, the weighted feature vectors are fused and the similarities between images are obtained and used as the measure bases to implement the image retrieval. The experiments show that the tangible results of image retrieval are realized and the average recall rate and the average precision rate are improved.Then, the formation of cable fault and the common fault types was described in this paper. The cable fault system model was build using the MATLAB software, and several common cable fault signal images were simulated from the model. Such as single-phase open circuit fault, single-phase short-circuit ground, the two phase short circuit ground, three-phase short-circuit voltage waveform and so on.Finally, the image texture features are first applied in cable signal fault recognition in the early cable fault on line recognition problems. R parameter based on gray-level co-occurrence matrix is defined, the method of texture features R parameter extraction is given, and the cable fault signal recognition method based on R parameters of texture features image processing is proposed. The recognition of several common fault patterns can be achieved correctly and effectively proved by the experiments.