发明名称 Image detection method and device
摘要 A method and device for detecting images are used to improve a detection rate and accuracy rate of the image moving target detection, so that moving targets also can be accurately detected in a scenario with complicated illumination changes. The method for detecting the images includes: collecting an image, performing moving target detection on the image by using a preset mixed space-time background model, and determining a target image; wherein, the mixed space-time background model is obtained in advance by modeling according to a grayscale change trend of the image.
申请公布号 US9361702(B2) 申请公布日期 2016.06.07
申请号 US201214127555 申请日期 2012.06.27
申请人 ZTE Corporation 发明人 Fu Qiang;Li Bo;Zhang Zhenwei;Chen Qimei;He Jun;Jiang Dengbiao
分类号 G06T7/20 主分类号 G06T7/20
代理机构 Ling and Yang Intellectual Property 代理人 Wu Ling;Yang Stephen;Ling and Yang Intellectual Property
主权项 1. A method for detecting images, comprising: modeling to obtain a mixed space-time background model according to a grayscale change trend of an image; and collecting the image, performing moving target detection on the image by using the mixed space-time background model to determine a moving target image; wherein said modeling to obtain a mixed space-time background model according to a grayscale change trend of an image comprises: determining a predicted pixel value based on exponential smoothing and illumination changes; integrating the predicted pixel value and a Center Symmetric Local Binary Pattern (CSLBP) operator to obtain a Space Time-Center Symmetric-Local Binary Pattern (ST-CSLBP) operator; calculating histograms corresponding to all pixels based on the ST-CSLBP operator; and constructing a mixed space-time background model of the histograms corresponding to all the pixels based on the ST-CSLBP operator; wherein the predicted pixel value is determined by using the following formula:zt=mt+1-ββ⁢rt-1rt=β⁡(xt-xt-1)+(1-β)⁢rt-1mt=β⁢⁢xt+(1-β)⁢mt-1; wherein zt represents a predicted pixel value of a pixel at time t, mt represents an estimated state value of the pixel at time t, β represents a smoothing factor, rt−1 represents a grayscale change trend of the pixel at time t−1, rt represents a grayscale change trend of the pixel at time t, xt represents an observed pixel value of the pixel at time t, and xt−1 represents an observed pixel value of the pixel at time t−1.
地址 Shenzhen, Guangdong Province CN