Thresholding and Hybrid CLAHE-HE for Chicken Egg Embryo Segmentation

This study aims to detect the presence of chicken egg embryos using digital image processing. The detection of chicken egg embryos was processed by image segmentation using the thresholding method. The implementation of this method needs to be supplemented with preprocessing phase using methods such as cropping, grayscaling, intensity adjustment, and histogram equalization (HE). The critical emphasis in this research is on image enhancement using a hybrid of Contrast Limited Additive Histogram Equalization (CLAHE) and HE. Based on comparing CLAHE, HE, hybrid CLAHE-HE, or HE-CLAHE, the optimal result used as input for the segmentation process is hybrid CLAHE-HE. The segmentation results using the thresholding method and hybrid CLAHE-HE showed clear and characteristic pictures of chicken egg embryos incomplete detection. Other methods, such as Otsu and k-means, also showed clear and characteristic pictures of chicken egg embryos. However, in terms of processing speed, the thresholding method was faster than the Otsu and k-means methods. The speed comparisons are thresholding⁶
seconds). These experiments show that our proposed method (thresholding and hybrid CLAHE-HE) can divide/segment embryonic eggs with significant embryonic characteristics and patterns.