Hand-foot-and-mouth disease detection method, device and system based on thermal infrared and RGB double shooting

Abstract

The embodiment of the invention discloses a hand-foot-and-mouth disease detection method, device and system based on thermal infrared and RGB double shooting. The hand-foot-and-mouth disease detectionmethod comprises the following steps: acquiring temperature through a thermal infrared camera; acquiring face and hand images through an RGB camera; performing spatial association on the temperatureand the human face to obtain an association result; using an improved YOLO-V3 focus detection network to detect the face and hand images; combining the temperature, the association result, the face detection result and the hand detection result to carry out joint judgment to obtain a judgment result; and triggering whether to alarm according to a judgment result. For the hand-foot-and-mouth disease detection method, body temperature measurement and hand-foot-and-mouth detection can be completed at the same time; all detection and calculation show a detection result within 1s at a local end with a neural network accelerator; whether the body temperature and the skin have papules, macular papules or herpes lesions or not is judged in a combined mode; the hand-foot-and-mouth disease detectionaccuracy is improved; an abnormal alarm is further set; and a worker can be reminded to treat a detected person in time.

Description

technical field [0001] The invention relates to the technical field of computer vision and disease prevention, in particular to a detection method, device and system for hand, foot and mouth disease based on thermal infrared and RGB dual-camera. Background technique [0002] Hand, foot and mouth disease is a common infectious disease in children around 5 years old. At present, most kindergartens use health doctors to measure body temperature, and then check whether there are pimples, maculopapules or herpes on the palms, backs of hands, tongue, face and mouth to determine whether there is hand, foot and mouth disease, but this will easily lead to physician fatigue and suspected cases cannot be recorded in time , and the passage time of each child entering the kindergarten is relatively slow in 5-10 seconds. Moreover, doctors and every child are contacted and tested, and there is a risk of acting as an intermediate host to spread the virus. [0003] For the detection of HFM...

AI Technical Summary

Problems solved by technology

[0007] (1) The current technology cannot automatically detect hand, foot and mouth symptoms and measure body temperature at the same time

[0008] (2) Adopting the front-end collection and server identification method to pass through the inspection is slow, which is greatly affected by the network bandwidth and server performance

[0009] (3) The picture of the suspected disease cannot match the identity of the person to be tested, which is not conducive to retrospective analysis and case tracking

[0010] (4) The design of the neural network is not appropriate, resulting in too many restrictions on the input image and many missed detections of small targets

[0011] (5) Lack of localized integrated detection, investigation and record reporting scheme, which is not conducive to prevention and control linkage and timely reporting

Images