In remote areas, wireless multimedia sensor networks (WMSNs) have limited energy,\nand the data processing of wildlife monitoring images always suffers from energy consumption\nlimitations. Generally, only part of each wildlife image is valuable. Therefore, the above mentioned\nissue could be avoided by transmitting the target area. Inspired by this transport strategy, in this\npaper, we propose an image extraction method with a low computational complexity, which can\nbe adapted to extract the target area (i.e., the animal) and its background area according to the\ncharacteristics of the image pixels. Specifically, we first reconstruct a color space model via a CIELUV\n(LUV) color space framework to extract the color parameters. Next, according to the importance of\nthe Hermite polynomial, a Hermite filter is utilized to extract the texture features, which ensures\nthe accuracy of the split extraction of wildlife images. Then, an adaptive mean-shift algorithm is\nintroduced to cluster texture features and color space information, realizing the extraction of the\nforeground area in the monitoring image. To verify the performance of the algorithm, a demonstration\nof the extraction of field-captured wildlife images is presented. Further, we conduct a comparative\nexperiment with N-cuts (N-cuts), the existing aggregating super-pixels (SAS) algorithm, and the\nhistogram contrast saliency detection (HCS) algorithm. A comparison of the results shows that the\nproposed algorithm for monitoring image target area extraction increased the average pixel accuracy\nby 11.25%, 5.46%, and 10.39%, respectively; improved the relative limit measurement accuracy by\n1.83%, 5.28%, and 12.05%, respectively; and increased the average mean intersection over the union\nby 7.09%, 14.96%, and 19.14%, respectively.
Loading....