Online detection device for flange glass bug based on machine vision

The utility model discloses a float glass defect online detection device based on machine vision. A light source adopts a red light emitting diode, a light source cabinet is positioned below the glass measured, and the slit at the top part of the light source cabinet is vertical to the traveling direction of the glass. A camera group is positioned above the glass measured, and a client package, an industrial switch and a signal filter are arranged in a control cabinet. The client group is provided with n client machines with image acquisition cards, the client machines are respectively connected with the cameras in the camera group, the signal filter is respectively connected with a photoelectric encoder and each client machine, and the photoelectric encoder is arranged on the glass measured. A marking machine is positioned above the glass measured, the marking machine is connected with a server, the server controls the operation of the marking machine, the defect information is recorded into a product defect database, and a statistical form is output. The device of the utility model can detect in real time, display and store the defects generated during the production process of the float glass, and can realize the accurate classification and recognition to the defects, such as inclusion, air bubble and optical distortion, etc.