Seven-segment Display Automatic Detection and Interpretation System using CNN-GO

In industrial plants, machines and measuring instruments are displayed through a seven-segment display. The machine is operating for more than 20 hours/day. The operator cannot check measurement data or status all the time. In addition, human error affects overall performance and time. This research aims to develop an embedded system for interpreting data from seven segment displays through online image processing. CNN (Convolution Neural Network) is applied in the detection and interpretation process. This paper proposes seven-segment display automatic detection and interpretation system using CNN-GO. An IoT device takes the photo of the measuring instrument's seven-segment display and sends the image to the server. The server interprets an image to numerical data using CNN-GO. Grayscale and Overlapping scanning are applied to increase the accuracy of detection and interpretation of numerical data.