Applied Sciences

Research

17 pages, 2642 KiB

Open AccessArticle

Ensemble Learning of Multiple Deep CNNs Using Accuracy-Based Weighted Voting for ASL Recognition

by Ying Ma, Tianpei Xu, Seokbung Han and Kangchul Kim

Appl. Sci. 2022, 12(22), 11766; https://doi.org/10.3390/app122211766 - 19 Nov 2022

Cited by 2 | Viewed by 1597

More than four million people worldwide suffer from hearing loss. Recently, new CNNs and deep ensemble-learning technologies have brought promising opportunities to the image-recognition field, so many studies aiming to recognize American Sign Language (ASL) have been conducted to help these people express [...] Read more.

More than four million people worldwide suffer from hearing loss. Recently, new CNNs and deep ensemble-learning technologies have brought promising opportunities to the image-recognition field, so many studies aiming to recognize American Sign Language (ASL) have been conducted to help these people express their thoughts. This paper proposes an ASL Recognition System using Multiple deep CNNs and accuracy-based weighted voting (ARS-MA) composed of three parts: data preprocessing, feature extraction, and classification. Ensemble learning using multiple deep CNNs based on LeNet, AlexNet, VGGNet, GoogleNet, and ResNet were set up for the feature extraction and their results were used to create three new datasets for classification. The proposed accuracy-based weighted voting (AWV) algorithm and four existing machine algorithms were compared for the classification. Two parameters,

α

and

λ

, are introduced to increase the accuracy and reduce the testing time in AWV. The experimental results show that the proposed ARS-MA achieved 98.83% and 98.79% accuracy on the ASL Alphabet and ASLA datasets, respectively. Full article

(This article belongs to the Special Issue Advances in Applied Signal and Image Processing Technology)

► Show Figures

Figure 1

16 pages, 7894 KiB

Open AccessArticle

Utilizing Mask R-CNN for Solid-Volume Food Instance Segmentation and Calorie Estimation

by Yanyan Dai, Subin Park and Kidong Lee

Appl. Sci. 2022, 12(21), 10938; https://doi.org/10.3390/app122110938 - 28 Oct 2022

Cited by 3 | Viewed by 2219

Abstract

To prevent or deal with chronic diseases, using a smart device, automatically classifying food categories, estimating food volume and nutrients, and recording dietary intake are considered challenges. In this work, a novel real-time vision-based method for solid-volume food instance segmentation and calorie estimation [...] Read more.

To prevent or deal with chronic diseases, using a smart device, automatically classifying food categories, estimating food volume and nutrients, and recording dietary intake are considered challenges. In this work, a novel real-time vision-based method for solid-volume food instance segmentation and calorie estimation is utilized, based on Mask R-CNN. In order to address the proposed method in real life, distinguishing it from other methods which use 3D LiDARs or RGB-D cameras, this work applies RGB images to train the model and uses a simple monocular camera to test the result. Gimbap is selected as an example of solid-volume food to show the utilization of the proposed method. Firstly, in order to improve detection accuracy, the proposed labeling approach for the Gimbap image datasets is introduced, based on the posture of Gimbap in plates. Secondly, an optimized model to detect Gimbap is created by fine-tuning Mask R-CNN architecture. After training, the model reaches AP (0.5 IoU) of 88.13% for Gimbap1 and AP (0.5 IoU) of 82.72% for Gimbap2. mAP (0.5 IoU) of 85.43% is achieved. Thirdly, a novel calorie estimation approach is proposed, combining the calibration result and the Gimbap instance segmentation result. In the fourth section, it is also shown how to extend the calorie estimation approach to be used in any solid-volume food, such as pizza, cake, burger, fried shrimp, oranges, and donuts. Compared with other food calorie estimation methods based on Faster R-CNN, the proposed method uses mask information and considers unseen food. Therefore, the method in this paper outperforms the accuracy of food segmentation and calorie estimation. The effectiveness of the proposed approaches is proven. Full article

(This article belongs to the Special Issue Advances in Applied Signal and Image Processing Technology)

► Show Figures

Figure 1

15 pages, 4324 KiB

Open AccessArticle

Reconstruction Optimization Algorithm of 3D Temperature Distribution Based on Tucker Decomposition

by Zhaoyu Liu, Shi Liu, Yaofang Zhang and Pengbo Yao

Appl. Sci. 2022, 12(21), 10814; https://doi.org/10.3390/app122110814 - 25 Oct 2022

Viewed by 1154

Abstract

For the purpose of solving the large temperature field reconstruction error caused by different measuring point arrangements and the problem that the prior dataset cannot be built due to data loss or distortion in actual measurement, a three-dimensional temperature profile reconstruction optimization algorithm [...] Read more.

For the purpose of solving the large temperature field reconstruction error caused by different measuring point arrangements and the problem that the prior dataset cannot be built due to data loss or distortion in actual measurement, a three-dimensional temperature profile reconstruction optimization algorithm is proposed to repair the empirical dataset and optimize the arrangement of temperature measuring points based on Tucker decomposition, the minimum condition number method, the greedy algorithm, and the hill climbing algorithm. We used the Tucker decomposition algorithm to repair the missing data and obtain the complete prior dataset and the core tensor. By optimizing the dimension of the core tensor and the number and position of the measuring points calculated by the minimum condition number method, the greedy algorithm, and the mountain climbing algorithm, the real-time three-dimensional distribution of the temperature field is reconstructed. The results show that the Tucker decomposition optimization algorithm could accurately complete the prior dataset, and compared with the original algorithm, the proposed optimal placement algorithm improves the reconstruction accuracy by more than 20%. At the same time, the algorithm has strong robustness and anti-noise, and the relative error is less than 4.0% and 6.0% with different signal-to-noise ratios. It indicates that the proposed method can solve the problem of building an empirical dataset and 3D temperature distribution reconstruction more accurately and stably in industry. Full article

(This article belongs to the Special Issue Advances in Applied Signal and Image Processing Technology)

► Show Figures

Figure 1

14 pages, 1050 KiB

Open AccessArticle

Image Noise Reduction by Means of Bootstrapping-Based Fuzzy Numbers

by Reza Ghasemi, Samuel Morillas, Ahmad Nezakati and Mohammadreza Rabiei

Appl. Sci. 2022, 12(19), 9445; https://doi.org/10.3390/app12199445 - 21 Sep 2022

Viewed by 1029

Abstract

Removing or reducing noise in color images is one of the most important functions of image processing, which is used in many sciences. In many cases, nonlinear methods significantly reduce the noise in the image and are widely used today. One of these [...] Read more.

Removing or reducing noise in color images is one of the most important functions of image processing, which is used in many sciences. In many cases, nonlinear methods significantly reduce the noise in the image and are widely used today. One of these methods is the use of fuzzy logic. In this paper, we want to introduce a fuzzy filter by using the fuzzy metric for fuzzy sets. For this purpose, we define fuzzy color pixels by using the mean of neighborhoods. Due to the noise in the image, we use the bootstrap resampling method to reduce the effect of outliers. The concept of the strong law of large numbers for the bootstrap mean in fuzzy metric space helps us to use the resampling method. Full article

(This article belongs to the Special Issue Advances in Applied Signal and Image Processing Technology)

► Show Figures

Figure 1

24 pages, 13873 KiB

Open AccessArticle

Preprocessing Acoustic Emission Signal of Broken Wires in Bridge Cables

by Guangming Li, Zhen Zhao, Yaohan Li, Chun-Yin Li and Chi-Chung Lee

Appl. Sci. 2022, 12(13), 6727; https://doi.org/10.3390/app12136727 - 2 Jul 2022

Cited by 5 | Viewed by 1712

Abstract

Bridges, especially cable-stayed bridges, play an important role in modern transportation systems. The safety status of bridge cables, as an important component of cable-stayed bridges, determines the health status of the entire bridge. As a non-destructive real-time detection technology, acoustic emission has the [...] Read more.

Bridges, especially cable-stayed bridges, play an important role in modern transportation systems. The safety status of bridge cables, as an important component of cable-stayed bridges, determines the health status of the entire bridge. As a non-destructive real-time detection technology, acoustic emission has the advantages of high detection efficiency and low cost. This paper focuses on the issue that a large amount of data are generated during the process of health monitoring of bridge cables. A novel acoustic emission signal segmentation algorithm is proposed with the aim to facilitate the extraction of acoustic emission signal characteristics. The proposed algorithm can save data storage space efficiently. Moreover, it can be adapted to different working conditions according to the adjustment of parameters in order to accurately screen out the target acoustic emission signal. Through the acoustic emission signal acquisition experiments of three bridges, the characteristics of the noise signal in the acquisition process are extracted. A comprehensive analysis of the signal in the time domain, frequency domain and time-frequency domain is carried out. The noise signal filtering parameter thresholds are proposed according to the analysis results. Full article

(This article belongs to the Special Issue Advances in Applied Signal and Image Processing Technology)

► Show Figures

Figure 1

15 pages, 8848 KiB

Open AccessArticle

Image Style Transfer via Multi-Style Geometry Warping

by Ioana Alexandru, Constantin Nicula, Cristian Prodan, Răzvan-Paul Rotaru, Mihai-Lucian Voncilă, Nicolae Tarbă and Costin-Anton Boiangiu

Appl. Sci. 2022, 12(12), 6055; https://doi.org/10.3390/app12126055 - 14 Jun 2022

Cited by 3 | Viewed by 1807

Abstract

Style transfer of an image has been receiving attention from the scientific community since its inception in 2015. This topic is characterized by an accelerated process of innovation; it has been defined by techniques that blend content and style, first covering only textural [...] Read more.

Style transfer of an image has been receiving attention from the scientific community since its inception in 2015. This topic is characterized by an accelerated process of innovation; it has been defined by techniques that blend content and style, first covering only textural details, and subsequently incorporating compositional features. The results of such techniques has had a significant impact on our understanding of the inner workings of Convolutional Neural Networks. Recent research has shown an increasing interest in the geometric deformation of images, since it is a defining trait for different artists, and in various art styles, that previous methods failed to account for. However, current approaches are limited to matching class deformations in order to obtain adequate outputs. This paper solves these limitations by combining previous works in a framework that can perform geometric deformation on images using different styles from multiple artists by building an architecture that uses multiple style images and one content image as input. The proposed framework uses a combination of various other existing frameworks in order to obtain a more intriguing artistic result. The framework first detects objects of interest from various classes inside the image and assigns them a bounding box, before matching each detected object image found in a bounding box with a similar style image and performing warping on each of them on the basis of these similarities. Next, the algorithm blends back together all the warped images so they are placed in a similar position as the initial image, and style transfer is finally applied between the merged warped images and a different chosen image. We manage to obtain stylistically pleasing results that were possible to generate in a reasonable amount of time, compared to other existing methods. Full article

(This article belongs to the Special Issue Advances in Applied Signal and Image Processing Technology)

► Show Figures

Figure 1

18 pages, 12632 KiB

Open AccessArticle

5D Gauss Map Perspective to Image Encryption with Transfer Learning Validation

by Sharad Salunke, Bharti Ahuja, Mohammad Farukh Hashmi, Venkatadri Marriboyina and Neeraj Dhanraj Bokde

Appl. Sci. 2022, 12(11), 5321; https://doi.org/10.3390/app12115321 - 24 May 2022

Cited by 7 | Viewed by 2067

Abstract

Encryption of visual data is a requirement of the modern day. This is obvious and greatly required due to widespread use of digital communication mediums, their wide range of applications, and phishing activities. Chaos approaches have been shown to be extremely effective among [...] Read more.

Encryption of visual data is a requirement of the modern day. This is obvious and greatly required due to widespread use of digital communication mediums, their wide range of applications, and phishing activities. Chaos approaches have been shown to be extremely effective among many encryption methods. However, low-dimensional chaotic schemes are characterized by restricted system components and fundamental structures. As a result, chaotic signal estimation algorithms may be utilized to anticipate system properties and their initial values to breach the security. High-dimensional chaotic maps on the other hand, have exceptional chaotic behavior and complex structure because of increased number of system parameters. Therefore, to overcome the shortcomings of the lower order chaotic map, this paper proposes a 5D Gauss Map for image encryption for the first time. The work presented here is an expansion of the Gauss Map’s current 1D form. The performance of the stated work is evaluated using some of the most important metrics as well as the different attacks in the field. In addition to traditional and well-established metrics such as PSNR, MSE, SSIM, Information Entropy, NPCR, UACI, and Correlation Coefficient that have been used to validate encryption schemes, classification accuracy is also verified using transfer learning. The simulation was done on the MATLAB platform, and the classification accuracy after the encryption-decryption process is compared. Full article

(This article belongs to the Special Issue Advances in Applied Signal and Image Processing Technology)

► Show Figures

Figure 1

24 pages, 1989 KiB

Open AccessFeature PaperArticle

Experimental Assessment of Feature Extraction Techniques Applied to the Identification of Properties of Common Objects, Using a Radar System

by José Francisco Díez-Pastor, Pedro Latorre-Carmona, José Luis Garrido-Labrador, José Miguel Ramírez-Sanz and Juan J. Rodríguez

Appl. Sci. 2021, 11(15), 6745; https://doi.org/10.3390/app11156745 - 22 Jul 2021

Viewed by 1773

Abstract

Radar technology has evolved considerably in the last few decades. There are many areas where radar systems are applied, including air traffic control in airports, ocean surveillance, and research systems, to cite a few. Other types of sensors have recently appeared, which allow [...] Read more.

Radar technology has evolved considerably in the last few decades. There are many areas where radar systems are applied, including air traffic control in airports, ocean surveillance, and research systems, to cite a few. Other types of sensors have recently appeared, which allow tracking sub-millimeter motion with high speed and accuracy rates. These millimeter-wave radars are giving rise to myriad new applications, from the recognition of the material close objects are made, to the recognition of hand gestures. They have also been recently used to identify how a person interacts with digital devices through the physical environment (Tangible User Interfaces, TUIs). In this case, the radar is used to detect the orientation, movement, or distance from the objects to the user’s hands or the digital device. This paper presents a thoughtful comparative analysis of different feature extraction techniques and classification strategies applied on a series of datasets that cover problems such as the identification of materials, element counting, or determining the orientation and distance of objects to the sensor. The results outperform previous works using these datasets, especially when the accuracy was lowest, showing the benefits feature extraction techniques have on classification performance. Full article

(This article belongs to the Special Issue Advances in Applied Signal and Image Processing Technology)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Advances in Applied Signal and Image Processing Technology

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Published Papers (8 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI