Vision Enhancement using Image Processing

This study compares two popular image preprocessing methods: dark channel and bilateral filtering. They are evaluated on a set of low-light images and compared based on visual quality metrics. Dark channel performs better for removing haze and improving contrast, while bilateral filtering is more effective for preserving details and reducing noise. To evaluate the performance of the methods in a practical application, they are applied to dataset provided. Both methods can effectively enhance nighttime driving images, but bilateral filtering produces more visually pleasing results with fewer halo artifacts. The study can give some idea for selection of appropriate techniques for specific applications, such as enhancing visibility in foggy scenarios