Target detection method and device, equipment, medium and computer program product

The invention provides a target detection method, device and equipment, a medium and a computer program product, and the method comprises the steps: inputting a to-be-detected image into a deep residual network in a training stage to obtain an extraction feature, inputting the extraction feature into a preset feature fusion module, and presetting a target horizontal anchor frame on each layer of output feature of the feature fusion module; calculating a matching degree between a real target frame corresponding to the to-be-detected image and each preset target horizontal anchor frame; screening positive and negative samples from each target horizontal anchor frame according to the matching degree, inputting the positive and negative samples into a detection regression network, and training the detection regression network through a preset regression loss function; and in a test stage, inputting the output of the feature fusion module corresponding to the to-be-detected image into the trained detection regress