Object Detection and Segmentation in Indian Flat Bread Chapati Using AI Models

The present research evaluates the implementation of the YOLOv11n model, which stands as one of the best industry-level object detection algorithms, and the application of the SAM2 model, an advanced segmentation algorithm for the digital identification of Indian flatbread, commonly known as chapati. The research demonstrates how the enhanced features and real-time processing of the YOLOv11n system detect and classify chapati images under multiple environmental conditions. The main goal exploits SAM2's advanced ability to divide intricate shapes and textures to separate the chapati pictures from multiple backgrounds and lighting variations. Researchers used high-resolution chapati images under different lighting circumstances with diverse backgrounds for the training and validation process. The identification accuracy of chapatis using YOLOv11n reached notable levels, which established both its effectiveness and speedy operation. The SAM2 performance evaluation method incorporates metrics related to segmentation precision combined with system execution speed and its ability to handle different conditions. This study demonstrates how YOLOv11 object identification and SAM2 segmentation are combined to simplify the industrial food processing operations and automate the food production systems within the food industry framework.