Abstract: In view of problems of lower real-time performance and higher false rate of existing detection methods of electric bucket teeth, an electric bucket teeth missing detection method based on machine vision was put forward. The method uses infrared thermal imager to collect bucket image, adopts template matching principle to accurately locate target area of bucket teeth under complicated background, and uses frame differential method to realize motion detection of bucket teeth. On the basis of location of the target area, the bucket teeth are segmented and extracted combined with location relation of the bucket teeth line area and the teeth line structure characteristics, so as to detect missing bucket teeth according to self-adaptive threshold. The experimental results show that the method can effectively realize real-time on-line, rapid and accurate detection of the missing teeth of the electric bucket, detection accuracy is more than 90%, which provides an effective solution for electric bucket teeth missing detection.