Dynamic ultrasonic video-based focus area real-time segmentation device

The invention relates to a focus area real-time segmentation device based on a dynamic ultrasonic video. The focus area real-time segmentation device comprises an image acquisition module used for acquiring a current frame of a video stream ultrasonic image; and the segmentation module is used for inputting the current frame into an optimized neural network to obtain a segmentation result of the focus area, and the optimized neural network is obtained through linear operation integration and precision calibration. According to the invention, the lesion area of the video stream ultrasonic image can be segmented continuously and accurately in real time. 本发明涉及一种基于动态超声视频的病灶区域实时分割装置，包括：图像获取模块：用于获取视频流超声图像的当前帧；分割模块：用于将所述当前帧输入优化后的神经网络来得到病灶区域的分割结果，其中，所述优化后的神经网络通过线性运算整合和精度校准得到。本发明能够连续、准确且实时分割视频流超声图像的病灶区域。