Fabric-Defect-Identification-System

Generate the file retrained_graph.pb(Refer in web) using Tensorflow library and training data-set(about 250 images) of non-defective fabric images. The conveyor belt system has the fabric on it. Connect the conveyor belt's motors to motor-driver and the motor-driver to arduino. Power the motor-driver from power-supply. Connect arduino and camera to PC via USB.

Upload the motor_and_cam.ino file to arduino.

Using the file retrained_graph.pb...Run the code python motor_cam_2_pattern_no_pat.py using the command: python motor_cam_2_pattern_no_pat.py

NOTE: If any usb problem arises while running the motor_cam_2_pattern_no_pat code, execute these two commands in terminal and try again...

1)sudo usermod -a -G dialout "USERNAME OF THE SYSTEM"
2)sudo chmod a+rw /dev/ttyACM0

Refer the ppt and https://www.youtube.com/watch?v=hiOQAwpi-XY for more details.