search

experiencor / keras-yolo3

Training and Detecting Objects with YOLO3
MIT License
1.6k stars 861 forks source link

how to crop detected bounding box from output image #254

Open ghost opened 4 years ago

ghost commented 4 years ago

Please, help me. I am currently doing final year project. Car License Plate Recognition is the main part of the project. I need to crop the detected car license plate from the output image in order to use OCR to recognize car license number.

AhmedHessuin commented 4 years ago

this is so easy. in bbox.py before the module draw rectangle on the image. you can take the coordinates (xmin,ymin,xmax,ymax) and insert it in list. thus you will have every coordinate of every output , in for loop you can crop from the original image as new images.

ghost commented 4 years ago

thank you. I have done it and it worked.