In this paper, a smart contrast enhancement technique based on conventional histogram equalization (HE) algorithm is proposed. This dynamic histogram equalization (DHE) technique takes control over the effect of traditional HE so that it performs the enhancement of an image without making any loss of details in it. DHE partitions the image histogram based on local minima and assigns specific gray level ranges for each partition before equalizing them separately. These partitions further go though a repartitioning test to ensure the absence of any dominating portions. This method outperforms other present approaches by enhancing the contrast well without introducing severe side effects, such as washed out appearance, checkerboard effects etc., or undesirable artifacts.

A Dynamic Histogram Equalization for Image Contrast Enhancement

Image inpainting refers to the process of restoring missing or damaged areas in an image. This field of research has been very active over recent years, boosted by numerous applications: restoring images from scratches or text overlays, loss concealment in a context of impaired image transmission, object removal in a context of editing, or disocclusion in image-based rendering (IBR) of viewpoints different from those captured by the cameras. Although earlier work dealing with disocclusion has been published in [1], the term inpainting first appeared in [2] by analogy with a process used in art restoration.

Image Inpainting : Overview and Recent Advances

A review of smart home applications based on Internet of Things

MPEG-4 is the multimedia standard for combining interactivity, natural and synthetic digital video, audio and computer-graphics Typical applications are: internet, video conferencing, mobile videophones, multimedia cooperative work, teleteaching and games With MPEG-4 the next step from block-based video (ISO/IEC MPEG-1, MPEG-2, CCITT H261, ITU-T H263) to arbitrarily-shaped visual objects is taken This significant step demands a new methodology for system analysis and design to meet the considerably higher flexibility of MPEG-4 Motion estimation is a central part of MPEG-1/2/4 and H261/H263 video compression standards and has attracted much attention in research and industry, for the following reasons: it is computationally the most demanding algorithm of a video encoder (about 60-80% of the total computation time), it has a high impact on the visual quality of a video encoder, and it is not standardized, thus being open to competition Algorithms, Complexity Analysis, and VLSI Architectures for MPEG-4 Motion Estimation covers in detail every single step in the design of a MPEG-1/2/4 or H261/H263 compliant video encoder: Fast motion estimation algorithms Complexity analysis tools Detailed complexity analysis of a software implementation of MPEG-4 video Complexity and visual quality analysis of fast motion estimation algorithms within MPEG-4 Design space on motion estimation VLSI architectures Detailed VLSI design examples of (1) a high throughput and (2) a low-power MPEG-4 motion estimator Algorithms, Complexity Analysis and VLSI Architectures for MPEG-4 Motion Estimation is an important introduction to numerous algorithmic, architectural and system design aspects of the multimedia standard MPEG-4 As such, all researchers, students and practitioners working in image processing, video coding or system and VLSI design will find this book of interest

Algorithms, Complexity Analysis and VLSI Architectures for MPEG-4 Motion Estimation

Neural Networks (NN) are a family of models for a broad range of emerging machine learning and pattern recondition applications. NN techniques are conventionally executed on general-purpose processors (such as CPU and GPGPU), which are usually not energy-efficient since they invest excessive hardware resources to flexibly support various workloads. Consequently, application-specific hardware accelerators for neural networks have been proposed recently to improve the energy-efficiency. However, such accelerators were designed for a small set of NN techniques sharing similar computational patterns, and they adopt complex and informative instructions (control signals) directly corresponding to high-level functional blocks of an NN (such as layers), or even an NN as a whole. Although straightforward and easy-to-implement for a limited set of similar NN techniques, the lack of agility in the instruction set prevents such accelerator designs from supporting a variety of different NN techniques with sufficient flexibility and efficiency.In this paper, we propose a novel domain-specific Instruction Set Architecture (ISA) for NN accelerators, called Cambricon, which is a load-store architecture that integrates scalar, vector, matrix, logical, data transfer, and control instructions, based on a comprehensive analysis of existing NN techniques. Our evaluation over a total of ten representative yet distinct NN techniques have demonstrated that Cambricon exhibits strong descriptive capacity over a broad range of NN techniques, and provides higher code density than general-purpose ISAs such as ×86, MIPS, and GPGPU. Compared to the latest state-of-the-art NN accelerator design DaDianNao [5] (which can only accommodate 3 types of NN techniques), our Cambricon-based accelerator prototype implemented in TSMC 65nm technology incurs only negligible latency/power/area overheads, with a versatile coverage of 10 different NN benchmarks.

Cambricon: an instruction set architecture for neural networks

Mammography is the first process to acquire breast imaging data. Approved radiologists screen the mammogram images for detecting suspicious lesions, e.g., mass, calcification, asymmetry. However, image screening is time-consuming and is a subjective decision-making process. In this paper, we propose an asymmetry recognition scheme with various convolutional neural network (CNN) architectures. The proposed scheme not only determines whether a pair of mammogram images is asymmetry, but it indicates that which image asymmetry occurs in.

Asymmetry Recognition of Mammogram Images Based on Convolutional Neural Network

A method for picture-based barcode decoding is provided herein. The method for picture-based barcode decoding includes the following steps: capturing all or part of a picture as a captured image; normalizing the captured image for generating a transformed image; calculating a mean color value of a plurality of blocks of the transformed image for generating a decoded data; performing an inverse error correction on the decoded data for generating an original data bitstream; transforming the original data bitstream into an original data; and outputting the original data with an output device.

Method and device for picture-based barcode decoding

We propose a scheme to detect blurry edge and compute sharpness measure. The key is color line model which models a pixel color is the linear mixture of two dominant colors in small patch. K-means clustering with principal component analysis is exploited to classify colors into two groups. Dominant color is defined as average color of group. Subsequently, weighting of color is computed. In this work, weighting is available for blurry edge detection, and color difference between synthesized color and original color is available for sharpness measure. The proposed scheme has better performance in blurry edge detection and sharpness measurement.

Blurry edge detection and sharpness measure using color line model with K-means clustering

A method for picture-based barcode encoding and decoding is provided herein. The method for picture-based barcode encoding includes: transforming an original data into an original data bitstream; performing an error correction on the original data bitstream for translating the original data bitstream into an error corrected bitstream; selecting all or part of the picture as an encoded area; calculating a data storage capacity of the encoded area; adjusting a size of the error corrected bitstream or a size of the encoded area for equalizing a data storage capacity of an encoded data bitstream and the data storage capacity of the encoded area; and adjusting a pixel value of the encoded area according to the encoded data bitstream.

Method and device for picture-based barcode encoding and decoding

This paper presents a method to localize and to recover the tampered areas in the watermarked image simultaneously. First, the proposed watermarking algorithm aims to embed a bi-watermark, which is the binary halftone of the downscaled host image, into the host image. While the watermarked image is tampered unauthorizedly, the extracted bi-watermark can be exploited to localize the tampered areas automatically without initial bi-watermark. Subsequently, Gaussian lowpass filter and quadratic programming are used to restore the bi-watermark to gray-scale image for inverse halftoning procedure. Eventually, the restored gray-scale image is used to recover the tampered areas.

Yi-Chong Zeng

Papers

Asymmetry Recognition of Mammogram Images Based on Convolutional Neural Network

Method and device for picture-based barcode decoding

Blurry edge detection and sharpness measure using color line model with K-means clustering

Method and device for picture-based barcode encoding and decoding

Joint Bi-Watermarking and Halftoning Technique Capability for Both Tampered Areas Localization and Recovery of Still Image