cv::dnn::NMSBoxes
时间: 2023-09-22 18:12:10 浏览: 39
cv::dnn::NMSBoxes is a function in the OpenCV library that performs non-maximum suppression on the detected object bounding boxes. Non-maximum suppression is a technique used in object detection algorithms to eliminate overlapping bounding boxes and retain only the most confident detections. This function takes as input a vector of bounding boxes, their confidence scores, a threshold value, and a parameter for the maximum number of output boxes. It returns a vector of indices corresponding to the selected bounding boxes after non-maximum suppression. This function is commonly used in deep learning-based object detection models to post-process the output and improve the accuracy of the detections.