Edge detection is one of the significant properties of many computer vision systems in image recognition, enhancement, compression, restoration, and so on. Though many research results were published, researchers still try to develop a more efficient edge detection algorithm that meets the current needs. We propose an image pre-processing based edge detection technique using canny edge detector. The main problem of canny edge detector is that it cannot identify the edges which are slightly vague due to the Gaussian smoothing. In this study, we solve the problem using the histogram processing of an image before using the canny edge detector. The result demonstrates that the proposed approach gives better outcomes compared to the state-of-the-art edge detectors.