Implementação de algoritmo de processamento digital de imagens para segmentação de ovos de galinha

Abstract

This work presents the development of an algorithm for automatic segmentation of eggs in images using digital image processing techniques. The goal is to develop an efficient and accurate method for identifying and separating eggs in a series of images that contain eggs arranged in rows and columns. The developed method goes through several steps, including image pre-processing, edge detection, egg identification, segmentation and analysis of the results. With the help of the OpenCV, NumPy and Matplotlib libraries, it was possible to implement the algorithm effectively. The results obtained at the end demonstrated 100 percent accuracy in identifying and sorting eggs for all test cases. However, it is important to highlight that there is still room for improvement, such as the integration of machine learning techniques to increase segmentation accuracy. The development of this work contributes as a complement to a research project developed at the Computer Vision and Data Recovery Laboratory at UFU, so that it is feasible to automate the stage of segmenting eggs from images, which was previously carried out semi-automatically.