Experienced OpenCV programmer needed to create script to allow Raspberry Pi with a picamera to identify two different types of objects, rotate the image if necessary to straighten it out, decode barcode data found in the image, and save a new image that is named with part of the barcode data and another label for the type of image it detected.
For example, when a plastic tray is put underneath the camera with an order invoice, products, receipts, labels... I would like the script to be able to see that it is indeed an order about to be packaged, take a high resolution picture, attempt to straighten the image, extract data from visible barcodes and then if an order # is found in the barcode data, save the image as $[login to view URL]
If a sealed package is identified (like a package in a box with a shipping label, or an envelope with a shipping label), take a high resolution picture, straighten the image, extract the barcode data visible and again save the image with a name like $[login to view URL]
if no barcode data is found but the object is identified save is as either <datetimestamp>[login to view URL] or <datetimestamp>[login to view URL]