85%。The locust detection and count method based on image processing was developed. The method was used to extract the number of locusts from color videos captured by a novel locust plague forecasting system with a super-low altitude helicopter. The color of locust is sandy beige, and the locusts’ habitat is mainly composed of green leaves. Therefore, the region of locust was segmented at first from the habitat according to the color feature of the hue space. Then some single locusts were detected by the area and compactness of the segmented locust area, and the mean area of them was used to compute the number of locusts for the characteristic that the infestation locusts have generally same size. The experimental results showed that the precision of the number of locusts achieved 85%.
參考文獻(xiàn)
相似文獻(xiàn)
引證文獻(xiàn)
引用本文
毛文華,鄭永軍,苑嚴(yán)偉,張小超.基于色度和形態(tài)特征的蝗蟲(chóng)信息提取技術(shù)[J].農(nóng)業(yè)機(jī)械學(xué)報(bào),2008,39(9):104-107.[J]. Transactions of the Chinese Society for Agricultural Machinery,2008,39(9):104-107.