CNN-Based Suppression of False Contour and Color Distortion in Bit-Depth Enhancement.

Changmeng Peng,Xiaofeng Li,Luting Cai,Zhizhong Fu,Xiaoyang Huang,Jin Xu

doi:10.3390/s21020416

Abstract

It is a challenge to transmit and store the massive visual data generated in the Visual Internet of Things (VIoT), so the compression of the visual data is of great significance to VIoT. Compressing bit-depth of images is very cost-effective to reduce the large volume of visual data. However, compressing the bit-depth will introduce false contour, and color distortion would occur in the reconstructed image. False contour and color distortion suppression become critical issues of the bit-depth enhancement in VIoT. To solve these problems, a Bit-depth Enhancement method with AUTO-encoder-like structure (BE-AUTO) is proposed in this paper. Based on the convolution-combined-with-deconvolution codec and global skip of BE-AUTO, this method can effectively suppress false contour and color distortion, thus achieving the state-of-the-art objective metric and visual quality in the reconstructed images, making it more suitable for bit-depth enhancement in VIoT.

Highlights

Visual sensors can provide richer and more intuitive information compared to other types of sensors
We find color distortion in some of the result images and that this distortion is caused by out-of-bounds values
When we check the output of the Bit-depth Enhancement method with AUTO-encoder-like structure (BE-AUTO)-base, we find that the value range of the output image exceeds the expected interval [−1, 1]

Summary

Introduction

Visual sensors can provide richer and more intuitive information compared to other types of sensors. They are critical Perception Front Ends (PFEs) of VIoT, and have been used in many scenarios, such as security surveillance, person identification, image retrieval, and telemedicine [1,2]. Compared with other types of signals, such as temperature sensor signals, visual signals have a huge amount of data. It is a great challenge to transmit and store these visual data. Each PFE in VIoT can only supply the compact memory space, limited energy and computing resources. The visual signals perceived by PFE must be sent to the cloud platform with massive storage space and rich computing resources [1]. Based on the existing video data encoding and decoding framework, it is very necessary to do proper preprocessing before the visual data is compressed [4]

Methods

Results

Discussion

Conclusion