Researchers at the Indian Institute of Technology Madras have developed a technique to to reduce the impact of haze on images captured by surveillance cameras. Apart from helping police solve crimes, it can also be applied to self-driving autonomous vehicles to enable efficient and safe navigation.
The image quality of CCTV cameras are usually poor in foggy or hazy conditions. A team of researchers led by A.N. Rajagopalan, professor in the Electrical Engineering Department, developed a technique to de-haze such images, the results of which were published in the journal IEEE Transactions on Image Processing.
The haziness in a scene, as captured by a camera, has much to do with how said camera perceives the object or scene. The recorded image should ideally reveal the original colour contrast and brightness of the scene.
However, the medium between the object and camera can decrease the amount of light reaching the camera. The particles in the atmosphere reflect ambient light, which gets added to the directly transmitted component and produces a hazy effect in the captured image. When there is haze or fog, the suspended particles in the air block reflected light, causing a shift in colours and poor contrast. This also sometimes creates a white veil that results in bad picture quality.
Mr. Rajagopalan and Srimanta Mandal, a postdoctoral fellow at the institute, devised a post-processing technique to mitigate the effect of haziness. The scientists modelled portions of haze in the image from the standpoint of the image formation process. They worked to correct the decrease in intensity of light and change in image colour due to fog, which were the aspects responsible for the image’s haziness.
The research team was also able to handle haziness in images taken during night or even underwater.