Deep Concatenated Residual Networks for Improving Quality of Halftoning-Based BTC Decoded Image

Heri Prasetyo; Alim Wicaksono Hari Prayuda; Chih-Hsien Hsia; Jing-Ming Guo

doi:10.3390/jimaging7020013

Deep Concatenated Residual Networks for Improving Quality of Halftoning-Based BTC Decoded Image

Journal of Imaging ◽

10.3390/jimaging7020013 ◽

2021 ◽

Vol 7 (2) ◽

pp. 13

Author(s):

Heri Prasetyo ◽

Alim Wicaksono Hari Prayuda ◽

Chih-Hsien Hsia ◽

Jing-Ming Guo

Keyword(s):

Simple Technique ◽

Impulsive Noise ◽

Visual Observation ◽

Compression Technique ◽

Block Truncation Coding ◽

Residual Learning ◽

Ill Posed ◽

Unpleasant Feeling ◽

Learning Frameworks

This paper presents a simple technique for improving the quality of the halftoning-based block truncation coding (H-BTC) decoded image. The H-BTC is an image compression technique inspired from typical block truncation coding (BTC). The H-BTC yields a better decoded image compared to that of the classical BTC scheme under human visual observation. However, the impulsive noise commonly appears on the H-BTC decoded image. It induces an unpleasant feeling while one observes this decoded image. Thus, the proposed method presented in this paper aims to suppress the occurring impulsive noise by exploiting a deep learning approach. This process can be regarded as an ill-posed inverse imaging problem, in which the solution candidates of a given problem can be extremely huge and undetermined. The proposed method utilizes the convolutional neural networks (CNN) and residual learning frameworks to solve the aforementioned problem. These frameworks effectively reduce the impulsive noise occurrence, and at the same time, it improves the quality of H-BTC decoded images. The experimental results show the effectiveness of the proposed method in terms of subjective and objective measurements.

Download Full-text

Metaheuristic techniques comparison to optimize robotic end-effector behavior and its workspace

International Journal of Advanced Robotic Systems ◽

10.1177/1729881418801132 ◽

2018 ◽

Vol 15 (5) ◽

pp. 172988141880113

Author(s):

Miguel Angel Funes Lora ◽

Edgar Alfredo Portilla-Flores ◽

Raul Rivera Blas ◽

Emmanuel Alejandro Merchán Cruz ◽

Manuel Faraón Carbajal Romero

Keyword(s):

Numerical Technique ◽

Cuckoo Search ◽

Optimization Methods ◽

Metaheuristic Algorithms ◽

End Effector ◽

Bee Colony ◽

Ill Posed ◽

Metaheuristic Techniques

Many robots are dedicated to replicate trajectories recorded manually; the recorded trajectories may contain singularities, which occur when positions and/or orientations are not achievable by the robot. The optimization of those trajectories is a complex issue and classical optimization methods present a deficient performance when solving them. Metaheuristics are well-known methodologies for solving hard engineering problems. In this case, they are applied to obtain alternative trajectories that pass as closely as possible to the original one, reorienting the end-effector and displacing its position to avoid the singularities caused by limitations of inverse kinematics equations, the task, and the workspace. In this article, alternative solutions for an ill-posed problem concerning the behavior of the robotic end-effector position and orientation are proposed using metaheuristic algorithms such as cuckoo search, differential evolution, and modified artificial bee colony. The case study for this work considers a three-revolute robot (3R), whose trajectories can contain or not singularities, and an optimization problem is defined to minimize the objective function that represents the error of the alternative trajectories. A tournament in combination with a constant of proportionality allows the metaheuristics to modify the gradual orientation and position of the robot when a singularity is present. Consequently, the procedure selects from all the possible elbow-configurations the best that fits the trajectory. A classical numerical technique, Newton’s method, is used to compare the results of the applied metaheuristics, to evaluate their quality. The results of this implementation indicate that metaheuristic algorithms are an efficient tool to solve the problem of optimizing the end-effector behavior, because of the quality of the alternative trajectory produced.

Download Full-text

X-rays Image reconstruction using Proximal Algorithm and adapted TV Regularization

MATEC Web of Conferences ◽

10.1051/matecconf/202134801011 ◽

2021 ◽

Vol 348 ◽

pp. 01011

Author(s):

Aicha Allag ◽

Redouane Drai ◽

Tarek Boutkedjirt ◽

Abdessalam Benammar ◽

Wahiba Djerir

Keyword(s):

Structural Information ◽

X Rays ◽

Reconstruction Algorithms ◽

Reconstruction Problem ◽

Tv Regularization ◽

Object Based ◽

Suggested Technique ◽

Reconstruction Image ◽

Ill Posed

Computed tomography (CT) aims to reconstruct an internal distribution of an object based on projection measurements. In the case of a limited number of projections, the reconstruction problem becomes significantly ill-posed. Practically, reconstruction algorithms play a crucial role in overcoming this problem. In the case of missing or incomplete data, and in order to improve the quality of the reconstruction image, the choice of a sparse regularisation by adding l1 norm is needed. The reconstruction problem is then based on using proximal operators. We are interested in the Douglas-Rachford method and employ total variation (TV) regularization. An efficient technique based on these concepts is proposed in this study. The primary goal is to achieve high-quality reconstructed images in terms of PSNR parameter and relative error. The numerical simulation results demonstrate that the suggested technique minimizes noise and artifacts while preserving structural information. The results are encouraging and indicate the effectiveness of the proposed strategy.

Download Full-text

Video Environmental Assessment of the Administrative Regions Within Urbgeosystem Territories

Man and Environment. Issues of Neoecology ◽

10.26565/1992-4224-2019-31-07 ◽

2019 ◽

Cited By ~ 1

Keyword(s):

Housing Stock ◽

Physiological State ◽

Video Recording ◽

Visual Fields ◽

Urban Ecosystem ◽

Visual Observation ◽

Visual Environment ◽

Visual Elements ◽

The Brain

Purpose. Assess the visual environment of the Novobavarskiy district within Kharkiv urban ecosystem. Methods. Field-based visual observation, photophixation and video recording, statistical. Results. We used a five-point scale to assess the "attractiveness" of territories and objects regarding the psycho-physiological state of a person to determine the quality of the visual environment of Novobavarskiy district, Kharkiv. As a result of video-environmental studies, it was discovered that there are homogeneous and aggressive fields in the visual environment on the territory of Kharkiv. In most cases the aesthetics of the district architecture has a neutral character, bacause a significant number of buildings does not have a variety of visual elements, so as homogeneous and aggressive areas can be found in great variety. It is determined that in the district there are 36% of comfortable visual fields, 13% are homogeneous and 51% are aggressive visual fields. Recently, there has been positive dynamics in the formation of a comfortable visual environment of the district's housing stock. When building new houses and renovating old buildings, different colours for facades are used. The interior is filled with attractive children's playgrounds and green spaces which transforms the aggressive and homogeneous fields of the visual environment into a comfortable one. Conclusions. The situation shows that it is necessary to create a comfortable visual environment that is not represented sufficiently in this district. In this case, there is a real threat to the physiological functions of the brain regarding the perception of information about the visual environment. And in the future it is also necessary to solve these problems using technologies that have been tested and implemented successfully in European countries.

Download Full-text

High Resolution Imaging of Compact Radio Sources by Gridding with Regularization

International Astronomical Union Colloquium ◽

10.1017/s0252921100046182 ◽

1998 ◽

Vol 164 ◽

pp. 411-412 ◽

Cited By ~ 1

Author(s):

S. F. Likhachev ◽

R. M. Hjellming

Keyword(s):

High Resolution ◽

Image Reconstruction ◽

Radio Source ◽

Radio Sources ◽

High Resolution Imaging ◽

Traditional Methods ◽

Compact Radio Source ◽

Ill Posed ◽

Resolution Imaging

AbstractThe problem of VLBI image reconstruction is a classical example of an ill-posed problem. A new procedure of gridding with regularization has been developed. This procedure was used in traditional methods (CLEAN, Hybrid) to improve the quality of compact radio source images. A few sources (GRO J1655–40, RY Scuti and Cyg X-1), observed with the VLA and VLBA, were processed with this procedure.

Download Full-text

Convergence Rates for Exponentially Ill-Posed Inverse Problems with Impulsive Noise

SIAM Journal on Numerical Analysis ◽

10.1137/15m1022252 ◽

2016 ◽

Vol 54 (1) ◽

pp. 341-360 ◽

Cited By ~ 11

Author(s):

Claudia König ◽

Frank Werner ◽

Thorsten Hohage

Keyword(s):

Inverse Problems ◽

Convergence Rates ◽

Impulsive Noise ◽

Ill Posed

Download Full-text

Drying and sortation to improve green beans and cup quality of Wamena coffee

E3S Web of Conferences ◽

10.1051/e3sconf/202130603024 ◽

2021 ◽

Vol 306 ◽

pp. 03024

Author(s):

Adnan ◽

Martina Sri Lestari

Keyword(s):

Simple Technique ◽

Descriptive Analysis ◽

Quality Score ◽

Local Culture ◽

Green Coffee ◽

Green Beans ◽

Green Coffee Beans ◽

Coffee Beans ◽

Cup Quality

Drying and sortation are the most important steps to improve green coffee beans and cup quality. However, farmers very often neglect these steps. Therefore, a simple technique and soft approach are required to encourage farmers to implement drying and sortation technology. The study aim is to assess suitable drying and sortation technology to improve green coffee beans and cup quality to local culture in Jayawijaya Regency, Papua. The study was conducted using 2 factors; a. Combination of drying floor using a tarp and without sortation (DFWTS), b. Combination of drying tables and with sortation (DTWS). Drying tables were designed as two separate parts. The first part was the permanent tables, and the second part was removable boxes in dimension 80 x 80 cm located on top of the permanent tables. Descriptive analysis was conducted based on SNI 01-2907-2008 by the Indonesian Coffee and Cocoa Research Institute. The results show DTWS produce green coffee beans in compliance with SNI 01-2907-2008 at 4a grade, compared to DFWTS is rejected. Green coffee beans quality is likely to affect cup quality. DTWS obtain cup quality score 83.0 compare to DFWTS is 81.25. In conclusion, DTWS improve green coffee beans and cup quality.

Download Full-text

Preliminary Results of Integrating Tikhonov’s Regularization in Forward-Backward Time-Stepping Technique for Object Detection

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.833.170 ◽

2016 ◽

Vol 833 ◽

pp. 170-175 ◽

Cited By ~ 2

Author(s):

Andrew Sia Chew Chie ◽

Kismet Anak Hong Ping ◽

Yong Guang ◽

Ng Shi Wei ◽

Nordiana Rajaee

Keyword(s):

Image Reconstruction ◽

Inverse Scattering ◽

Time Domain ◽

Regularization Method ◽

Scattering Problem ◽

Inverse Scattering Problem ◽

Time Stepping ◽

Ill Posed ◽

Tikhonov’S Regularization

The inverse scattering in time domain known as Forward-Backward Time-Stepping (FBTS) technique is applied to determine the sizes, shape and location of the embedded objects. Tikhonov’s regularization method has been proposed in order to improve or solve the ill-posed of FBTS inverse scattering problem. The reconstructed results showed that FBTS technique can detect the presence of embedded objects. The reconstructed results of FBTS technique utilizing with the Tikhonov’s regularization method shown better results than the results only applied FBTS technique. Tikhonov’s regularization combined with FBTS technique to improve the quality of image reconstruction.

Download Full-text

Deep robust residual network for super-resolution of 2D fetal brain MRI

Scientific Reports ◽

10.1038/s41598-021-03979-1 ◽

2022 ◽

Vol 12 (1) ◽

Author(s):

Liyao Song ◽

Quan Wang ◽

Ting Liu ◽

Haiwei Li ◽

Jiancun Fan ◽

...

Keyword(s):

Spatial Resolution ◽

High Efficiency ◽

Brain Mri ◽

Fetal Brain ◽

Super Resolution ◽

Adult Brain ◽

Superior Performance ◽

Residual Learning ◽

2D Data

AbstractSpatial resolution is a key factor of quantitatively evaluating the quality of magnetic resonance imagery (MRI). Super-resolution (SR) approaches can improve its spatial resolution by reconstructing high-resolution (HR) images from low-resolution (LR) ones to meet clinical and scientific requirements. To increase the quality of brain MRI, we study a robust residual-learning SR network (RRLSRN) to generate a sharp HR brain image from an LR input. Due to the Charbonnier loss can handle outliers well, and Gradient Difference Loss (GDL) can sharpen an image, we combined the Charbonnier loss and GDL to improve the robustness of the model and enhance the texture information of SR results. Two MRI datasets of adult brain, Kirby 21 and NAMIC, were used to train and verify the effectiveness of our model. To further verify the generalizability and robustness of the proposed model, we collected eight clinical fetal brain MRI 2D data for evaluation. The experimental results have shown that the proposed deep residual-learning network achieved superior performance and high efficiency over other compared methods.

Download Full-text

An Effective Regularization Method for Image Super Resolution

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.219-220.1411 ◽

2011 ◽

Vol 219-220 ◽

pp. 1411-1414

Author(s):

En Wei Zheng ◽

Xian Jun Wang

Keyword(s):

Super Resolution ◽

License Plate ◽

Reconstruction Method ◽

Ill Posed ◽

High Resolution Images ◽

Image Super Resolution ◽

Sr Reconstruction ◽

Real Traffic ◽

Reconstruction Model

In this paper, we propose a new super resolution (SR) reconstruction method to handle license plate numbers of vehicles in real traffic videos. Recently, SR reconstruction shemes based on regularization have been demonstrated to be effective because SR reconstrction is an ill-posed problem. Working within this promising framework, the residual data (RD) term can be weighted according to the differences among the observed LR images in the SR reconstruction model. Moreover, L1 norm is used to measure the RD term in order to improve the robustness of our method. Experiments show the proposed method improves the subjective visual quality of the high resolution images.

Download Full-text