<p>Image edge detection is a crucial step in digital image processing that plays a pivotal role in various applications of computer vision. It involves identifying the abrupt changes in intensity or color that delineate boundaries between different regions in an image. Edge detection is used extensively in object recognition, image segmentation, medical imaging, autonomous vehicles, and robotics. Despite its importance, detecting edges accurately remains challenging due to noise, variations in image contrast, and complex image structures. In this research, we propose a novel approach to edge detection using the Butterfly-Optimized Adaptive Resonance Neural Model (BO-ARNM). BO-ARNM is an optimized neural network model that uses the parallel-learning capability of the Butterfly Optimization algorithm and the self-organizing properties of the Adaptive Resonance Theory (ART) neural network. Our approach uses the ART's vigilance parameter to increase the detection of edge pixels, enhancing overall edge detection accuracy. The performance of the BO-ARNM algorithm is compared with other previous edge detection algorithms such as SVM and CNN. The evaluation metrics used to assess the performance of the algorithm were accuracy, PSNR, SSIM, FAR and FRR. Experimental results on synthetic and real-world images demonstrate that the BO-ARNM achieves superior edge detection accuracy while also being more time-efficient than other compared methods. The proposed BO-ARNM approach to edge detection exhibits robustness against noise and variations in image contrast, making it suitable for real-world applications. The proposed BO-ARNM edge detection algorithm presents a novel contribution to the development of efficient and accurate edge detection systems that offer improved performance than other state-of-the-art methods. The results of this research have significant implications for the development of robust and efficient image processing methods and their applications in various fields.</p>