Abstract
Image intensifiers are used internationally as advanced military night-vision devices. They have better imaging performance in low-light-level conditions than CMOS/CCD. The intensified CMOS (ICMOS) was developed to satisfy the digital demand of image intensifiers. In order to make the ICMOS capable of color imaging in low-light-level conditions, a liquid-crystal tunable filter based color imaging ICMOS was developed. Due to the time-division color imaging scheme, motion artifacts may be introduced when a moving target is in the scene. To solve this problem, a deformable kernel prediction neural network (DKPNN) is proposed for joint denoising and motion artifact removal, and a data generation method which generates images with color-channel motion artifacts is also proposed to train the DKPNN. The results show that, compared with other denoising methods, the proposed DKPNN performed better both on generated noisy data and on real noisy data. Therefore, the proposed DKPNN is more suitable for color ICMOS denoising and motion artifact removal. A new exploration was made for low-light-level color imaging schemes.
Highlights
Since the color level that a human eye can distinguish is hundreds of times greater than the grayscale level [1], color imaging can make full use of the sensitivity of the human eye to color information, thereby improving its ability of scene understanding
intensified CMOS (ICMOS)/ICCD was developed by coupling the phosphor screen of the image intensifier to the CMOS/CCD sensor, thereby digitizing the output of the image intensifier
An liquid-crystal tunable filter (LCTF)-based color imaging ICMOS experimental system was set up in this article; due to the time-division color imaging scheme, if there were moving targets in the scene, color-channel motion artifacts were introduced in the final image output
Summary
Since the color level that a human eye can distinguish is hundreds of times greater than the grayscale level [1], color imaging can make full use of the sensitivity of the human eye to color information, thereby improving its ability of scene understanding. A data generation strategy for color-channel motion artifacts is presented, and the proposed DKPNN was trained using the generated data This network can simultaneously complete denoising and motion artifact removal. (1) A data generation strategy for color-channel motion artifacts; (2) A DKPNN which achieves joint denoising and motion artifact removal, suitable for color ICMOS/ICCD; (3) An LCTF-ICMOS low-light-level color imaging system with high integration. The remainder of this article is arranged as follows: Section 2 details each step of the proposed DKPNN for joint denoising and motion artifact removal; Section 3 describes the setup of the LCTF-ICMOS experimental imaging system; Section 4 presents the validation results of the proposed method and discusses its effectiveness; and Section 5 concludes the paper and discusses future works
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have