Detecting objects in images is crucial across a wide range of applications, including surveillance, autonomous navigation, augmented reality, and more. While AI-based approaches such as Convolutional Neural Networks (CNNs) have proven highly effective in object detection, in some industrial applications where the objects being recognized are confidential, it is difficult to train an AI for such tasks. On the other hand, feature-based approaches like SIFT, SURF, and ORB offer the capability to search any template but may struggle with complex visual variations. In this work, we introduce a novel edge-based object/scene recognition method. We propose that utilizing feature edges, instead of feature points, offers high performance under complex visual variations. Our primary contribution is a directional pixel voting descriptor based on image segments. Experimental results are promising; compared to previous approaches, ours demonstrates superior performance under complex visual variations, enabling real-time processing with embedded capabilities.