Images and bounding box annotations of sound and rotten cranberry fruit from a postharvest imaging platform

This data set provides the raw images and bounding box annotations from a postharvest berry imaging platform that was used to train an object detection model. Annotations are of fruit belonging to two classes: sound and rotten. Downstream image processing of detected and classified berries measures the percent fruit rot in a sample. Images were collected using a Nikon D7500 DSLR camera using manual shutter speed, focal length, f-stop, and ISO settings. The imaging platform was a custom-build LED lightbox and sample tray. The dataset contains 60 images with accompanying annotations in YOLOv8 format.