scholarly journals Deep Learning-Based Multinational Banknote Type and Fitness Classification with the Combined Images by Visible-Light Reflection and Infrared-Light Transmission Image Sensors

Sensors ◽  
2019 ◽  
Vol 19 (4) ◽  
pp. 792 ◽  
Author(s):  
Tuyen Pham ◽  
Dat Nguyen ◽  
Chanhum Park ◽  
Kang Park
Keyword(s):  
Deep Learning ◽  
Visible Light ◽  
Light Transmission ◽  
Infrared Light ◽  
Light Reflection ◽  
United States Dollar ◽  
Fitness Value ◽  
Classification Tasks ◽  
Combined Images

Automatic sorting of banknotes in payment facilities, such as automated payment machines or vending machines, consists of many tasks such as recognition of banknote type, classification of fitness for recirculation, and counterfeit detection. Previous studies addressing these problems have mostly reported separately on each of these classification tasks and for a specific type of currency only. In other words, there has been little research conducted considering a combination of these multiple tasks, such as classification of banknote denomination and fitness of banknotes, as well as considering a multinational currency condition of the method. To overcome this issue, we propose a multinational banknote type and fitness classification method that both recognizes the denomination and input direction of banknotes and determines whether the banknote is suitable for reuse or should be replaced by a new one. We also propose a method for estimating the fitness value of banknotes and the consistency of the estimation results among input trials of a banknote. Our method is based on a combination of infrared-light transmission and visible-light reflection images of the input banknote and uses deep-learning techniques with a convolutional neural network. The experimental results on a dataset composed of Indian rupee (INR), Korean won (KRW), and United States dollar (USD) banknote images with mixture of two and three fitness levels showed that the proposed method gives good performance in the combination condition of currency types and classification tasks.

scholarly journals Deep Learning-Based Multinational Banknote Fitness Classification with a Combination of Visible-Light Reflection and Infrared-Light Transmission Images

Symmetry ◽  
2018 ◽  
Vol 10 (10) ◽  
pp. 431 ◽  
Author(s):  
Tuyen Pham ◽  
Dat Nguyen ◽  
Jin Kang ◽  
Kang Park
Keyword(s):  
Visible Light ◽  
Light Transmission ◽  
The United States ◽  
Infrared Light ◽  
Light Reflection ◽  
Fitness Level ◽  
United States Dollar ◽  
Automated Teller Machines ◽  
Fitness Levels

The fitness classification of a banknote is important as it assesses the quality of banknotes in automated banknote sorting facilities, such as counting or automated teller machines. The popular approaches are primarily based on image processing, with banknote images acquired by various sensors. However, most of these methods assume that the currency type, denomination, and exposed direction of the banknote are known. In other words, not only is a pre-classification of the type of input banknote required, but in some cases, the type of currency is required to be manually selected. To address this problem, we propose a multinational banknote fitness-classification method that simultaneously determines the fitness level of a banknote from multiple countries. This is achieved without the pre-classification of input direction and denomination of the banknote, using visible-light reflection and infrared-light transmission images of banknotes, and a convolutional neural network. The experimental results on the combined banknote image database consisting of the Indian rupee and Korean won with three fitness levels, and the United States dollar with two fitness levels, show that the proposed method achieves better accuracy than other fitness classification methods.

scholarly journals Recognition of Banknote Fitness Based on a Fuzzy System Using Visible Light Reflection and Near-infrared Light Transmission Images

Sensors ◽  
2016 ◽  
Vol 16 (6) ◽  
pp. 863 ◽  
Author(s):  
Seung Kwon ◽  
Tuyen Pham ◽  
Kang Park ◽  
Dae Jeong ◽  
Sungsoo Yoon
Keyword(s):  
Visible Light ◽  
Fuzzy System ◽  
Near Infrared ◽  
Light Transmission ◽  
Infrared Light ◽  
Light Reflection ◽  
Near Infrared Light

Biomimetics in Energy Systems: Light Transmission in the Window Plant Fenestraria aurantiaca as Inspiration for New Solutions in the Technical World

2012 ◽  
Vol 84 ◽  
pp. 51-56 ◽  
Author(s):  
Immanuel Schäfer
Keyword(s):  
Solar Cells ◽  
Visible Light ◽  
Mobile Devices ◽  
Near Infrared ◽  
Light Transmission ◽  
Energy Systems ◽  
Surface Cleaning ◽  
Infrared Light ◽  
Direct Sunlight ◽  
Light Concentration

Fenestraria aurantiaca (also known as window plant) is a succulent with specialized adaptations to deal with heat, light and aridity. Fenestraria aurantiaca (F. a.) grows with most of its body under the sand. Just the top, with a light transparent surface – the window – on it, protrudes from the surface hence giving explanation to the plants name. Experiments with light, and detailed microscopy studies show the physical, biological and chemical capabilities of F. a. It was found that the window works as a lens, light from a 90 ° angle is directed into the plant. Thereby the window filters the light. Up to 90 % of the visible light is blocked; with rising wavelength the window gets more transparent until the near infrared light (1000 nm) where the transparency declines rapidly. But the parenchyma is up 90 % transparent. Based on those results the principles of the plant were defined, which are used for abstractions. Generally F.a. has four principles: light handling, surface cleaning, heat avoidance and water storing. Improvements founded on the inspiration of the window plant seem to be possible in photovoltaic systems, which have problems with overheating and also light concentration. An example solution called “buried solar cells” is presented. Another working field is the screen of mobile devices, where the clarity and readability suffers from direct sunlight. With the help from the methods displayed by F.a., there is an energy saving solution explained.

scholarly journals Improved Active Deep Learning for Semi-Supervised Classification of Hyperspectral Image

2021 ◽  
Vol 14 (1) ◽  
pp. 171
Author(s):  
Qingyan Wang ◽  
Meng Chen ◽  
Junping Zhang ◽  
Shouqiang Kang ◽  
Yujing Wang
Keyword(s):  
Deep Learning ◽  
Supervised Classification ◽  
Hyperspectral Image ◽  
Graph Algorithm ◽  
Hyperspectral Images ◽  
Training Set ◽  
Proposed Model ◽  
Deep Networks ◽  
Classification Tasks

Hyperspectral image (HSI) data classification often faces the problem of the scarcity of labeled samples, which is considered to be one of the major challenges in the field of remote sensing. Although active deep networks have been successfully applied in semi-supervised classification tasks to address this problem, their performance inevitably meets the bottleneck due to the limitation of labeling cost. To address the aforementioned issue, this paper proposes a semi-supervised classification method for hyperspectral images that improves active deep learning. Specifically, the proposed model introduces the random multi-graph algorithm and replaces the expert mark in active learning with the anchor graph algorithm, which can label a considerable amount of unlabeled data precisely and automatically. In this way, a large number of pseudo-labeling samples would be added to the training subsets such that the model could be fine-tuned and the generalization performance could be improved without extra efforts for data manual labeling. Experiments based on three standard HSIs demonstrate that the proposed model can get better performance than other conventional methods, and they also outperform other studied algorithms in the case of a small training set.

scholarly journals Theoretic analysis and experimental evaluation of the spectrum transmission coefficient of a multilayer photovoltaic vacuum glazing

2020 ◽  
Vol 15 (4) ◽  
pp. 574-582
Author(s):  
Qinghua Lv ◽  
Jiachen Cui ◽  
Hasila Jarimi ◽  
Hui Lv ◽  
Zhongsheng Zhai ◽  
...  
Keyword(s):  
Thin Film ◽  
Visible Light ◽  
Transmission Coefficient ◽  
Light Transmission ◽  
Theoretical Calculations ◽  
Solar Spectrum ◽  
Infrared Light ◽  
Calculation Error ◽  
Functional Coating ◽  
Vacuum Glazing

Abstract This paper introduces an innovative thin film PV vacuum glazing (PV-VG) technology. In addition to electricity generation, the PV-VG glazing can also reduce heat loss from the building in winter and reduce heat gain in summer. In building integrated photovoltaics application, optical characterization of the PV glazing is important in determining the solar ray transmission and thermal transfer process of the glazing. This paper discusses the optical properties of the PV-VG glazing by considering the different layers of the glazing unit that includes a self-cleaning glass, a thin film PV glass and a low-e vacuum glazing. Based on the optical transfer matrix, the transmission coefficients of different film layers were deduced. The theoretical calculations were then validated against the transmission coefficient experiment of the PV-VG using an EDTM SS2450 Solar Spectrum Meter. The calculation error of the transmission coefficient of the single-layer glazing is generally within 5%, the calculation error of the transmission coefficient of the integrated PV-VG glazing is about 6%. The results show that the average visible light transmission coefficient, the average infrared light transmission coefficient and the overall transmission coefficient of PV-VG glazing are 19%, 16% and 12%, respectively. The study is important to optimize the visible light transmission of the PV-VG glazing; the optical model obtained above lays a solid foundation for further study of transmission coefficient analysis of different functional coating of PV-VG glazing.

scholarly journals Automatic analysis of magnetograms for identification and classification of active regions using Deep Learning

2020 ◽  
Vol 12 (2) ◽  
pp. 67-79
Author(s):  
Letícia Sousa De Oliveira ◽  
André Leon Sampaio Gradvohl
Keyword(s):  
Deep Learning ◽  
Solar Flares ◽  
Active Regions ◽  
The Sun ◽  
Learning Models ◽  
Technological Systems ◽  
Complete Identification ◽  
Learning Technique ◽  
Classification Tasks

Some phenomena that occur in the Sun have consequences on Earth. Among these phenomena, solar flares release large amounts of radiation and energy that impact on Earth's life and technological systems. These flares usually come from sunspots, which derive from solar magnetic activities. One strategy to predict solar flares is to identify active regions, i. e., a group of sunspots with a high potential to cause solar flares. This paper reports the use of the deep learning technique to identify and classify active regions from magnetogram analysis. To achieve these tasks, we created a dataset with magnetograms and performed tests to choose the best deep learning models for the identification and classification of active regions. The results of the best models reached accuracies higher than 80% for both the identification and classification tasks. Based on these results, we implemented a system in Python to automate the complete identification and classification process, also reported in this paper.

scholarly journals Detection and classification of brain tumours from MRI images using faster R-CNN

2019 ◽  
Vol 13 (4) ◽  
pp. 337-342 ◽  
Author(s):  
Ercan Avşar ◽  
Kerem Salçin
Keyword(s):  
Deep Learning ◽  
Brain Tumours ◽  
Powerful Method ◽  
Resonance Imaging ◽  
Brain Images ◽  
Mri Brain ◽  
Classification Tasks ◽  
Magnetic Resonance Imaging Mri ◽  
Mri Brain Images

Magnetic resonance imaging (MRI) is a useful method for diagnosis of tumours in human brain. In this work, MRI images have been analysed to detect the regions containing tumour and classify these regions into three different tumour categories: meningioma, glioma, and pituitary. Deep learning is a relatively recent and powerful method for image classification tasks. Therefore, faster Region-based Convolutional Neural Networks (faster R-CNN), a deep learning method, has been utilized and implemented via TensorFlow library in this study. A publicly available dataset containing 3,064 MRI brain images (708 meningioma, 1426 glioma, 930 pituitary) of 233 patients has been used for training and testing of the classifier. It has been shown that faster R-CNN method can yield an accuracy of 91.66% which is higher than the related work using the same dataset.

Robust Design of High-Performance Transparent Solar Cell Structure Considering Natural Sunlight Illumination

2014 ◽  
Author(s):  
Shuangcheng Yu ◽  
Chen Wang ◽  
Cheng Sun ◽  
Wei Chen
Keyword(s):  
Solar Cells ◽  
Solar Cell ◽  
Visible Light ◽  
Robust Design ◽  
High Performance ◽  
Light Transmission ◽  
Cell Structure ◽  
Solar Spectrum ◽  
Infrared Light ◽  
Natural Sunlight

Transparent organic solar cells have recently attracted extensive interest considering their potential application for the power-generating window. By allowing the transmission of visible light while converting ultraviolet and near infrared light in the solar spectrum into electricity, transparent solar cells integrated into building facade provide a smart solution to the energy dilemma in urban area. However, current works mainly optimize the performance of solar cells for very limited incident condition, such as only considering normal incidence, which results in impractical designs for real applications. In this paper, we propose a robust design approach to achieve high-performance transparent solar cell based on a non-periodic photonic structure considering a broad range of incident conditions representing natural sunlight illumination. Statistical performances are used in the robust design formulation and efficient sampling techniques are further employed to improve the computational efficiency. The Pareto-optimal solutions are obtained according to the multicriteria preference with respect to maximizing the expected cell transparency and the expected energy conversion efficiency, and minimizing the performance variance due to the incidence variation. As one example of the optimized design, the absorbing efficiency of the solar cell could be up to 85% that of its opaque counterpart with 32% visible light transmission and 0.13% variation coefficient of transparency under the actual solar illumination and incident angles from 9am to 3pm. By using this design methodology, practically efficient cell structure is achieved based on the location and installation orientation of the solar window.

Sign in / Sign up

Export Citation Format

Share Document