Infrared target tracking method and device based on response graph fusion

The invention discloses an infrared target tracking method and device based on response graph fusion, and belongs to the technical field of infrared image processing. The method comprises the following steps: obtaining a target image corresponding to a tracking target in a starting frame infrared image and a position point corresponding to the target image, and respectively obtaining a normalizedgray feature and a direction gradient histogram feature of the starting frame target; according to a target position in the initial frame image, using a correlation filtering algorithm to train correlation filters of the two features respectively. Compared with a method for carrying out feature representation by utilizing a single feature and different feature series connection modes, the trackingprecision and the tracking success rate are improved to a certain extent, meanwhile, the mode can keep better tracking robustness for an infrared image sequence under the complex condition, and persistent robustness tracking