Multi-target forehead temperature measuring method capable of accurately positioning forehead

The invention discloses a multi-target forehead temperature measurement method capable of accurately positioning a forehead, and the method comprises the steps: obtaining a detectable visible light image face through the training of a feedback type neural network, and extracting a model of the coordinates of a key point at the lower edge of the forehead of the face; positioning a human face forehead area by utilizing the human face detection frame and forehead lower edge key point coordinates; judging whether the forehead is completely shielded or not in a human face forehead region by utilizing an image adaptive binarization theory, and selecting an effective temperature measurement region; and mapping the effective temperature measurement area of the visible light image to the infrared thermal image, and performing infrared data analysis on the effective temperature measurement area to obtain a temperature measurement result of the current face. 本发明公开一种额头精准定位的多目标额温测量方法，该方法包括：通过反馈式神经网络训练得到可检测可见光图像人脸、并提取人脸额头下