- Use of Python and OpenCV for acquisition and processing of images from multiple cameras. Application involves real-time processing of images to yield decision of pass/fail of object being inspected.
Time-constraint: Acquisition, processing and decision-making to occur within window of 0.3s
- Intel 8 Core i9-9900K CPU @ 3.60GHz PC
- Four USB monochrome global shutter cameras with max resolution of 800 x 400 pixels
- Position sensor
- ST microelectronics Nucleo microcontroller board.
Process flow ( all in real-time ):
-Position sensor is interfaced to Nucleo board and generates trigger when object arrives at correct position for cameras to acquire images.
- Nucleo board is connected to PC via USB and sends an ASCII character via Serial interface when it detects trigger from position sensor.
- PC must keep monitoring the Serial port on which the Nucleo is connected and listen for the correct ASCII character signalling for acquisition of images.
- Once character is received, PC needs to acquire and process images from all four cameras simultaneously and generate a pass/fail signal.
- This pass/fail signal is to be sent to the Nucleo board, which will in turn trigger actuation of a pneumatic actuator when a fail signal is received.
Note: Exact process flow may vary with changes needed as the project progresses.