The invention relates to a
pedestrian detection method and device with confrontation and defense capabilities, equipment, a storage medium and a program product. The method comprises the following steps: converting a target image into a grey-
scale map; adversarial
region detection is carried out on the grey-
scale map in a
frequency domain, adversarial region
mask maps of multiple groups of n-level frequency bands in the
frequency domain are obtained, and the adversarial region
mask maps of any group of any-level frequency bands are used for indicating the adversarial region of the grey-
scale map under the level
frequency band of the
frequency domain; fusing the confrontation area
mask graphs of the same group of n-level frequency bands in the plurality of groups of n-level frequency bands to obtain a plurality of target mask graphs, and based on each target mask graph, performing fuzzy
processing on the confrontation area in the target image to obtain a plurality of preprocessed images; and respectively carrying out
pedestrian detection on the plurality of preprocessed images through a
pedestrian detection model, and carrying out non-maximum suppression fusion on a detection result to obtain a target detection result. Therefore, various adversarial attacks can be effectively defended, so that the adversarial defense capability of
pedestrian detection is improved.