OPTIMIZING IMAGE RECOGNITION THROUGH A NOVEL SOFT COMPUTING ALGORITHM

Abstract

Soft computing is a critical component of artificial intelligence and a rapidly growing area of machine intelligence, with the primary goal of addressing complex and ambiguous situations. The primary challenge in developing new soft computing algorithms for image recognition is to balance high accuracy with computational efficiency and interpretability. In this study, a novel image recognition algorithm using soft computing techniques is proposed based on a neuro-fuzzy logic model. The proposed method combines the learning capabilities of artificial neural networks with the reasoning abilities of fuzzy logic. The proposed method undergoes training and testing on the Common Objects in Context (COCO) dataset. The images in the COCO dataset undergo preprocessing, including resizing and normalization, to ensure consistent size and pixel values for robust image recognition. The proposedmethod achieves high performance with 100% accuracy, 99.01% precision, 99.03% recall, and a 99.07% F1 score in image recognition. These results show that the combination of neuro-fuzzy logic is effective in image recognition tasks, indicating its suitability for applications requiring high accuracy and reliability.