The invention provides a target detection method and system based on a regression algorithm, and relates to the technical field of image detection, and the method comprises the steps: obtaining an on-site image of a to-be-detected device, and constructing a sample database; establishing a detection model of deep learning; carrying out grid division processing on the images in the sample database,and inputting the processed images into a detection model for iterative training until preset training accuracy is achieved; obtaining an image with to-be-detected equipment to form a verification set, and inputting the image in the verification set into the detection model for verification until preset verification accuracy is achieved; obtaining an image with a to-be-detected equipment defect toform a test set, and inputting the image in the test set into the detection model for testing until preset test accuracy is achieved; And inputting the to-be-detected image into the detection model for detection so as to output a target defect image. According to the target detection method and system provided by the embodiment of the invention, the target object can be intelligently screened, and the accuracy and timeliness of target detection are improved.