An automatic segmentation method based on fuzzy AGA for thermometer reading image

Aiming at the problems of low efficiency and difficult guarantee of reading precision in the conventional liquid thermometer verification, an automatic reading method is proposed. Based on the gray stretching transformation, the ratio of intraclass to interclass distance and the penalty coefficient are used as segmentation criterion, the segmentation threshold and the parameters of gray stretching function are optimized by the fuzzy AGA. Experiment and comparative analysis results show that the proposed method has good segmentation effect and provides a reliable guarantee for image reading recognition.