Quality of service (Qos) and priority aware models for adaptive efficient image retrieval in WSN using TBL routing with RLBP features

The problem of image retrieval in wireless sensor networks has been well studied. Towards image retrieval in WSN, various methods have been discussed earlier, but suffer to achieve higher performance in image mining. To improve the performance, an efficient QoS adaptive image mining technique has been presented in this paper. The method focused on the efficiency in image mining as well as achieving QoS in WSN. The image has been processed to remove the noise by applying Gabor filters. From the noise removed image, the local binary pattern has been generated at each region of the image to produce regional local binary pattern (RLBP). The RLBP feature extracted has been used to measure the similarity between various images. The method maintains in the taxonomy of various image classes, and each class has different features. The input query has been measured for its similarity towards various classes from taxonomy. According to the similarity a single class has been identified. Based on the class identified, a subset of nodes from WSN has been identified where the relevant Images are available. To reach the data nodes the method identifies the list of routes and estimates traffic bandwidth latency (TBL) support. Based on the value of TBL support a specific route has been selected to perform image retrieval. The RLBP feature generated has been transferred to the data nodes, where the method estimates RLBPS (RLBP similarity). According to the value of RLBP similarity, subsets of images have been selected and transmitted the source node. The method improves the performance of image mining in WSN with less complexity.

Automated summary

An efficient QoS adaptive image mining technique is proposed in this paper to improve the performance of image retrieval in wireless sensor networks. The method focuses on processing images, extracting features, identifying classes, selecting routes, and achieving higher performance in image mining in WSN.