Non-reference quality evaluation method for dynamic range of HDR video

The invention discloses an HDR video dynamic range no-reference quality evaluation method, which comprises the following steps of: calculating a boundary value of a bright part and a dark part of a picture according to photoelectric transmission characteristics of a current HDR signal and a current SDR signal, and extracting a bright part quantity and a dark part quantity; the method comprises the following steps: calculating spatial distribution characteristics and statistical information characteristics of the two images to obtain characteristic parameters used for evaluating the detail level of the image, preprocessing all the characteristic parameters, combining a subjective quality evaluation result, and utilizing a suitable machine learning model to realize a professional and effective no-reference quality evaluation model suitable for industry habits. According to the method, no-reference dynamic range quality evaluation can be carried out on the HDR video, the method is high in pertinence, and ideas a