Aiming at problems that efficiency of current outdour search and rescue is low and the effect of search and rescue is susceptible to environmental factors

such as nighttime

an outdoor search and rescue unmanned aerial vehicle（ UAV） using infrared vision is designed and manufactured. Search and rescue UAV uses onboard infrared camera FLIR TAU2 to collect images from the search and rescue area. Captured infrared images are transmitted to onboard TK1 for real-time detection of human targets in infrared image. UAV transmits the detected geographic location of the human target to the ground station

so that the search and rescue personnel can complete the rescue mission in time. In the stage of infrared human target detection

histogram of oriented gradient（ HOG） is used to characterize the infrared target

and the HOG feature classification detection is performed by linear support vector machine（ SVM） classifier. The test results verify that the designed search and rescue UAV can complete the outdoor search and rescue mission in a variety of real field scenes.