A novel development of algorithms for lens-less miniature camera is for revolutionising the areas of Augmented Reality (AR)/ Virtual Reality (VR), security, robotics and smart wearables. The researchers at the Indian Institute of Technology (IIT) Madras and Rice University, U.S. developed the algorithms.
Unlike the conventional cameras, the lens-less ones have numerous vision applications. The lens-less cameras normally gave blurred images, as they did not capture a sharp photograph of the scene. The algorithms developed now by the IIT,, Madras and Rice university helps in producing photo-realistic images from the blurred lensless capture.
Assistant professor at the Department of Electrical Engineering, IIT Kaushik Mitra and Ashok Veeraraghavan of Rice University led the researchers. Ashok Veeraraghavan in 2016 developed a lensless camera where a thin optical mask was placed just in front of the sensor at a distance of approximately 1 mm. However, the camera captured only blurred images.
Mitra stated that the existing algorithms were not helpful in getting good images from the blurred ones. They also gave low resolution and noisy mages. The professor said that they used Deep Learning to develop a reconstruction algorithm called ‘FlatNet’ for lensless cameras. This gave an improved result, he said and added that Flatnet was tested on various real and challenging scenarios. All the results that the researchers derived at showed that this gave better results in getting fine tuned images in lens-less cameras.
Lens-less imaging is a new technology and had great potential in solving imaging/vision problems. The findings were presented as a paper in the IEEE International Conference on Computer Vision and an extended version appeared in IEEE Transactions on Pattern Analysis and Machine Intelligence.
The National Science Foundation (NSF), Career and NSF Expedition, National Institutes of Health (NIH) Grant, US, Qualcomm Innovation Fellowship India 2020 are among the institutions that funded the project.