Visually impaired persons find it extremely difficult to perform cash transactions in outdoor environments. For assisting the visually challenged individuals, a YOLOv5 based deep neural network was designed to detect image based currency denominations. Thereby aid in completing the authentic transaction. The robust model was trained for images with currency notes in different backgrounds, multiple sides of the currency notes presented, notes around cluttered objects, notes near reflective surfaces, and blurred images of the currency notes. An annotated and augmented dataset of around 10,000 original images was created for developing the model. A pre-processing step to rescale all the images to 224 × 224 was applied to standardize the input to the neural network, and generalize the model for different platforms including single board computer and smartphones. The trained model showcased an average denomination recognition accuracy of 92.71% for an altogether different dataset. The trained model was deployed on Raspberry-Pi and Smartphone independently, and the outcome to detect the currency denomination from the image was successfully demonstrated. The model showcased adequate performance on different platforms, leading to the exploration of several other assistive applications based on the currency recognition model to improve the standard of living for visually challenged individuals.
{"title":"An Indian Currency Recognition Model for Assisting Visually Impaired Individuals","authors":"Madhav Pasumarthy, Rutvi Padhy, Raghuveer Yadav, Ganesh Subramaniam, Madhav Rao","doi":"10.1109/RASSE54974.2022.9989624","DOIUrl":"https://doi.org/10.1109/RASSE54974.2022.9989624","url":null,"abstract":"Visually impaired persons find it extremely difficult to perform cash transactions in outdoor environments. For assisting the visually challenged individuals, a YOLOv5 based deep neural network was designed to detect image based currency denominations. Thereby aid in completing the authentic transaction. The robust model was trained for images with currency notes in different backgrounds, multiple sides of the currency notes presented, notes around cluttered objects, notes near reflective surfaces, and blurred images of the currency notes. An annotated and augmented dataset of around 10,000 original images was created for developing the model. A pre-processing step to rescale all the images to 224 × 224 was applied to standardize the input to the neural network, and generalize the model for different platforms including single board computer and smartphones. The trained model showcased an average denomination recognition accuracy of 92.71% for an altogether different dataset. The trained model was deployed on Raspberry-Pi and Smartphone independently, and the outcome to detect the currency denomination from the image was successfully demonstrated. The model showcased adequate performance on different platforms, leading to the exploration of several other assistive applications based on the currency recognition model to improve the standard of living for visually challenged individuals.","PeriodicalId":382440,"journal":{"name":"2022 IEEE International Conference on Recent Advances in Systems Science and Engineering (RASSE)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125575814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-07DOI: 10.1109/RASSE54974.2022.9989708
You-Liang Xie, Xin-Rong Lin, Che-Wei Lin
This study proposed a hardware-friendly-CNN-based hardware implementation system for screening electrocardiogram (ECG) ectopic beat with an STM32 ARM microcontroller-based embedded artificial intelligence (AI) edge device. In single heartbeat classification, continuous wavelet transformation based SEmbedNet and simplified AlexNet/GoogLeNet with different pixels of 56/112 of input size were compared to choose the best and most efficient combination to implement in the hardware. Five classes of the ectopic beat are followed by the ANSI/AAMI EC57 guideline in the MIT-BIH arrhythmia database, including non-ectopic beat (N), supraventricular ectopic beat (S), ventricular ectopic beat (V), fusion beat (F), and unknown beat(Q). Besides, this study performed the model through k-fold cross-validation and choose the best model for hardware implementation. The classification result showed that using a 5-layer CNN (SEmbedNet) with an input image of pixel 56 could get better performance than an 8-layer CNN (simplified AlexNet) with a total accuracy of 99.89%. Besides, the combination of SEmbedNet with an input image size of pixel 56 and STM32 can achieve the benefits of 1.3s and 1.1 W per heartbeat in the classification task, and it only takes about 4 seconds. Moreover, a multiple-STM32 cross-validation platform was built to reduce the validation time. It can process more than a hundred thousand heartbeats in 6.4 hours.
{"title":"SEmbedNet: Hardware-Friendly CNN for Ectopic Beat Classification on STM32-Based Edge Device","authors":"You-Liang Xie, Xin-Rong Lin, Che-Wei Lin","doi":"10.1109/RASSE54974.2022.9989708","DOIUrl":"https://doi.org/10.1109/RASSE54974.2022.9989708","url":null,"abstract":"This study proposed a hardware-friendly-CNN-based hardware implementation system for screening electrocardiogram (ECG) ectopic beat with an STM32 ARM microcontroller-based embedded artificial intelligence (AI) edge device. In single heartbeat classification, continuous wavelet transformation based SEmbedNet and simplified AlexNet/GoogLeNet with different pixels of 56/112 of input size were compared to choose the best and most efficient combination to implement in the hardware. Five classes of the ectopic beat are followed by the ANSI/AAMI EC57 guideline in the MIT-BIH arrhythmia database, including non-ectopic beat (N), supraventricular ectopic beat (S), ventricular ectopic beat (V), fusion beat (F), and unknown beat(Q). Besides, this study performed the model through k-fold cross-validation and choose the best model for hardware implementation. The classification result showed that using a 5-layer CNN (SEmbedNet) with an input image of pixel 56 could get better performance than an 8-layer CNN (simplified AlexNet) with a total accuracy of 99.89%. Besides, the combination of SEmbedNet with an input image size of pixel 56 and STM32 can achieve the benefits of 1.3s and 1.1 W per heartbeat in the classification task, and it only takes about 4 seconds. Moreover, a multiple-STM32 cross-validation platform was built to reduce the validation time. It can process more than a hundred thousand heartbeats in 6.4 hours.","PeriodicalId":382440,"journal":{"name":"2022 IEEE International Conference on Recent Advances in Systems Science and Engineering (RASSE)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133656402","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-07DOI: 10.1109/RASSE54974.2022.10019733
Sirwe Saeedi, A. Fong, Ajay K. Gupta, Steve M. Carr
Machine commonsense reasoning (MCR) systems can significantly improve the way we interact with machines. MCR systems are therefore an important element in any human-centric applications. Recent advances in machine learning (ML) have enabled breakthroughs in MCR technologies. This paper aims to improve healthcare outcomes by making human-machine interactions more intuitive than before. It presents learning models developed for MCR. Specifically, it presents a critical analysis of state-of-the-art deep learning (DL) models for MCR. These include recurrent neural network (RNN), transfer learning (TL), and transformers. Transformers, in particular, have been found to be effective for a range of natural language processing (NLP) applications, including MCR. Based on the analysis, another contribution of this paper is to assemble useful MCR tools into an adaptable MCR toolbox. To ensure broad applicability, the toolbox can be customizable for different MCR applications. Our research focuses on two specific MCR applications: commonsense validation and commonsense explanation. The former concerns identifying statements that do not make commonsense. The latter aims at explaining the reason why a given statement does not make commonsense. The paper presents some preliminary results of applying elements of the assembled toolbox to the two MCR applications. These results indicate that it is possible to achieve near human performances using finely-tuned state-of-the-art DL methods for the two MCR applications.
{"title":"Improving Healthcare Outcomes with Learning Models for Machine Commonsense Reasoning Systems","authors":"Sirwe Saeedi, A. Fong, Ajay K. Gupta, Steve M. Carr","doi":"10.1109/RASSE54974.2022.10019733","DOIUrl":"https://doi.org/10.1109/RASSE54974.2022.10019733","url":null,"abstract":"Machine commonsense reasoning (MCR) systems can significantly improve the way we interact with machines. MCR systems are therefore an important element in any human-centric applications. Recent advances in machine learning (ML) have enabled breakthroughs in MCR technologies. This paper aims to improve healthcare outcomes by making human-machine interactions more intuitive than before. It presents learning models developed for MCR. Specifically, it presents a critical analysis of state-of-the-art deep learning (DL) models for MCR. These include recurrent neural network (RNN), transfer learning (TL), and transformers. Transformers, in particular, have been found to be effective for a range of natural language processing (NLP) applications, including MCR. Based on the analysis, another contribution of this paper is to assemble useful MCR tools into an adaptable MCR toolbox. To ensure broad applicability, the toolbox can be customizable for different MCR applications. Our research focuses on two specific MCR applications: commonsense validation and commonsense explanation. The former concerns identifying statements that do not make commonsense. The latter aims at explaining the reason why a given statement does not make commonsense. The paper presents some preliminary results of applying elements of the assembled toolbox to the two MCR applications. These results indicate that it is possible to achieve near human performances using finely-tuned state-of-the-art DL methods for the two MCR applications.","PeriodicalId":382440,"journal":{"name":"2022 IEEE International Conference on Recent Advances in Systems Science and Engineering (RASSE)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132338079","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-07DOI: 10.1109/RASSE54974.2022.9989597
Kai Lei, Ming-Yueh Ku, Shuenn-Yuh Lee
The purpose of the system is to establish a real-time arrhythmia recognition according to image, which can be easily implemented by hardware as artificial intelligence (AI) accelerator. Through the remote photoplethysmography (rPPG), the slight changes of the face are captured in a non-contact way, and the analysis of the AI algorithm can deduce the correlation between subtle change of the face and arrhythmia. The design of a conventional neural network has a large of multipliers and adders in the internal network, and multi-bit multipliers and adders usually cause a long critical path. Through the accelerated design based on the computer in memory (CIM) system, the time of transferring the data can be effectively reduced. While the high-precision network also has a lot of parameters, so we need to compress the model for the realization of hardware.
{"title":"Real-Time and Non-Contact Arrhythmia Recognition Algorithm for Hardware Implementation","authors":"Kai Lei, Ming-Yueh Ku, Shuenn-Yuh Lee","doi":"10.1109/RASSE54974.2022.9989597","DOIUrl":"https://doi.org/10.1109/RASSE54974.2022.9989597","url":null,"abstract":"The purpose of the system is to establish a real-time arrhythmia recognition according to image, which can be easily implemented by hardware as artificial intelligence (AI) accelerator. Through the remote photoplethysmography (rPPG), the slight changes of the face are captured in a non-contact way, and the analysis of the AI algorithm can deduce the correlation between subtle change of the face and arrhythmia. The design of a conventional neural network has a large of multipliers and adders in the internal network, and multi-bit multipliers and adders usually cause a long critical path. Through the accelerated design based on the computer in memory (CIM) system, the time of transferring the data can be effectively reduced. While the high-precision network also has a lot of parameters, so we need to compress the model for the realization of hardware.","PeriodicalId":382440,"journal":{"name":"2022 IEEE International Conference on Recent Advances in Systems Science and Engineering (RASSE)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134243056","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-07DOI: 10.1109/RASSE54974.2022.9989986
H. Yeh, Jun Zhou
Secured and robust wireless communication systems are critical in rapidly changing mobile fading channels. Further developing the 2x1 conjugate cancellation (CC), we proposed a 4x1 space-time (ST) orthogonal frequency division multiplexing (OFDM) system in conjunction with CC as a block coded two-path transmission scheme. This full-rate 4x1 STCCOFDM system alleviates the effect of inter-carrier interference (ICI) in mobile channels with an outstanding BER performance due to the high diversity order and two-path CC block coding scheme which offers high signal-to-ICI ratio. Both Walsh–Hadamard transform (WHT) and Zadoff-Chu transform (ZCT) are used as the orthogonal pre-coder to further improve bit error rate (BER) performance. Employing the unique pre-coder at the transmitter, the security is achieved at the user’s receiver terminal since the user must perform the inverse operation via a prior known pre-coder. By employing the same order M-ary modulation in transmission, this 4x1 pre-coded full-rate STCCOFDM systems offer an outstanding BER than that of the 4x1 pre-coded half-rate ST OFDM system in mobile channels with the same bandwidth efficiency. By using a higher order M-ary modulation in transmission, the 4x1 full-rate STCCOFDM systems offer an outstanding BER over the full-rate 4x1 ST OFDM in mobile environments with the same bandwidth efficiency. Simulations prove that this full-rate 4x1 STCCOFDM systems are robust to mobile channels and the architecture can be generalized to multiple receiver antennas in the fifth generation (5G) systems.
在快速变化的移动衰落信道中，安全可靠的无线通信系统至关重要。进一步发展2x1共轭抵消(CC)，我们提出了一个4x1时空(ST)正交频分复用(OFDM)系统结合CC作为块编码双路传输方案。该全速率4x1 stcccofdm系统缓解了移动信道中载波间干扰(ICI)的影响，由于其高分集阶和双径CC分组编码方案提供了高信噪比，具有出色的误码率性能。采用Walsh-Hadamard变换(WHT)和Zadoff-Chu变换(ZCT)作为正交预编码器，进一步提高了误码率(BER)性能。在发送端采用唯一的预编码器，由于用户必须通过先前已知的预编码器执行反向操作，因此在用户的接收端实现了安全性。通过在传输中采用相同阶数的M-ary调制，该4x1预编码全速率STCCOFDM系统在相同带宽效率的移动信道中比4x1预编码半速率ST OFDM系统提供了出色的BER。通过在传输中使用更高阶的M-ary调制，4x1全速率STCCOFDM系统在相同带宽效率的移动环境中比全速率4x1 ST OFDM提供出色的BER。仿真结果表明，该全速率4x1 STCCOFDM系统对移动信道具有鲁棒性，该架构可推广到5G系统中的多接收天线中。
{"title":"Design 4x1 Space-Time Conjugate Two-Path Full-Rate OFDM Systems","authors":"H. Yeh, Jun Zhou","doi":"10.1109/RASSE54974.2022.9989986","DOIUrl":"https://doi.org/10.1109/RASSE54974.2022.9989986","url":null,"abstract":"Secured and robust wireless communication systems are critical in rapidly changing mobile fading channels. Further developing the 2x1 conjugate cancellation (CC), we proposed a 4x1 space-time (ST) orthogonal frequency division multiplexing (OFDM) system in conjunction with CC as a block coded two-path transmission scheme. This full-rate 4x1 STCCOFDM system alleviates the effect of inter-carrier interference (ICI) in mobile channels with an outstanding BER performance due to the high diversity order and two-path CC block coding scheme which offers high signal-to-ICI ratio. Both Walsh–Hadamard transform (WHT) and Zadoff-Chu transform (ZCT) are used as the orthogonal pre-coder to further improve bit error rate (BER) performance. Employing the unique pre-coder at the transmitter, the security is achieved at the user’s receiver terminal since the user must perform the inverse operation via a prior known pre-coder. By employing the same order M-ary modulation in transmission, this 4x1 pre-coded full-rate STCCOFDM systems offer an outstanding BER than that of the 4x1 pre-coded half-rate ST OFDM system in mobile channels with the same bandwidth efficiency. By using a higher order M-ary modulation in transmission, the 4x1 full-rate STCCOFDM systems offer an outstanding BER over the full-rate 4x1 ST OFDM in mobile environments with the same bandwidth efficiency. Simulations prove that this full-rate 4x1 STCCOFDM systems are robust to mobile channels and the architecture can be generalized to multiple receiver antennas in the fifth generation (5G) systems.","PeriodicalId":382440,"journal":{"name":"2022 IEEE International Conference on Recent Advances in Systems Science and Engineering (RASSE)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114540443","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-07DOI: 10.1109/RASSE54974.2022.9989980
H. Yeh, Jun Zhou
In modern space and communication systems, it is desirable to have a low bit error rate (BER) in impulse noise (IN) and mobile fading channels. IN exists in varied transmission systems, such as digital video broadcasting-terrestrial (DVB-T), digital audio broadcasting (DAB), asymmetric digital subscriber line (ADSL), and the fifth generation (5G) networks. A 2x2 space-time parallel cancellation (STPC) transmission scheme joint with interleaved orthogonal frequency division multiplexing (IOFDM) system is presented in this paper to mitigate IN and mobile fading channel effects. The STPC OFDM system employs an architecture with two-path transmission to mitigate inter-carrier interference (ICI) in mobile fading channels. The interleaving process in IOFDM is employed for increasing mixed time and frequency domain diversity within the two-path STPC OFDM block. Hence, the STPC-IOFDM system characterizes the excellent mitigation of ICI due to the robustness of STPC in mobile fading channels while the interleaving process introduces time and frequency domain diversity to further effectively combat IN and frequency selective fading channels. It is demonstrated via simulations that the proposed STPC-IOFDM system is vigorous to numerous frequency selective environments with or without IN. Its BER performance outperforms ST-OFDM, ST-IOFDM, and STPC-OFDM systems in both IN and COST207 typical urban or bad urban mobile fading channels.
{"title":"Space-Time Parallel Cancellation Interleaved OFDM Systems in Impulsive Noise and Mobile Fading Channels","authors":"H. Yeh, Jun Zhou","doi":"10.1109/RASSE54974.2022.9989980","DOIUrl":"https://doi.org/10.1109/RASSE54974.2022.9989980","url":null,"abstract":"In modern space and communication systems, it is desirable to have a low bit error rate (BER) in impulse noise (IN) and mobile fading channels. IN exists in varied transmission systems, such as digital video broadcasting-terrestrial (DVB-T), digital audio broadcasting (DAB), asymmetric digital subscriber line (ADSL), and the fifth generation (5G) networks. A 2x2 space-time parallel cancellation (STPC) transmission scheme joint with interleaved orthogonal frequency division multiplexing (IOFDM) system is presented in this paper to mitigate IN and mobile fading channel effects. The STPC OFDM system employs an architecture with two-path transmission to mitigate inter-carrier interference (ICI) in mobile fading channels. The interleaving process in IOFDM is employed for increasing mixed time and frequency domain diversity within the two-path STPC OFDM block. Hence, the STPC-IOFDM system characterizes the excellent mitigation of ICI due to the robustness of STPC in mobile fading channels while the interleaving process introduces time and frequency domain diversity to further effectively combat IN and frequency selective fading channels. It is demonstrated via simulations that the proposed STPC-IOFDM system is vigorous to numerous frequency selective environments with or without IN. Its BER performance outperforms ST-OFDM, ST-IOFDM, and STPC-OFDM systems in both IN and COST207 typical urban or bad urban mobile fading channels.","PeriodicalId":382440,"journal":{"name":"2022 IEEE International Conference on Recent Advances in Systems Science and Engineering (RASSE)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129725066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-11-07DOI: 10.1109/RASSE54974.2022.9989848
Meng-Ling Wu, Jui-Hung Chang, P. Chung
Colorectal cancer (CRC) is the third most common malignancy and the second most deadly cancer. The most efficient way to determine CRC staging is to analyze whole slide digital pathology images; therefore, it is certainly important to ensure the accuracy of pathology slide analysis.We can obtain medical quantized data of pathological images by implementing deep learning methods. These methods not only can light pathologists’ load but also can provide accurate computing results.In this paper, we use U-2-NET as our backbone to perform Binary Image Segmentation on CRC pathology slides. CRC pathology slides have a variety of non-conforming shapes and colors which is an enormous challenge for detecting cancer areas. U-2-NET was originally used in the Salient Object Detection (SOD) task to find the most unique regions of human attention, which can be used to identify abnormal regions in pathological slices. Moreover, the RSU block of U-2-NET can handle long-term and short-term dependencies, which we believe helps maintain contextual information. With the large computational costs, U-2-NET is hard to implement for application. Our purposed method can use preprocessing, image-selecting mechanisms and transfer learning concepts to solve this problem.Our results show that the model trained with a small part of the data set and a modified small object function has the best results for Binary Image Segmentation of colorectal cancer pathology sections by U-2-NET, with the best IOU (0.77) and Dice Loss (0.83) compared with other models (MSRFCNN, FCN, SegNet, and Unet). Furthermore, after transferring learning using pre-trained weights from the SOD dataset, the results are improved compared to those of learning the network from scratch.
{"title":"Image Segmentation for Colorectal cancer histopathological images analysis","authors":"Meng-Ling Wu, Jui-Hung Chang, P. Chung","doi":"10.1109/RASSE54974.2022.9989848","DOIUrl":"https://doi.org/10.1109/RASSE54974.2022.9989848","url":null,"abstract":"Colorectal cancer (CRC) is the third most common malignancy and the second most deadly cancer. The most efficient way to determine CRC staging is to analyze whole slide digital pathology images; therefore, it is certainly important to ensure the accuracy of pathology slide analysis.We can obtain medical quantized data of pathological images by implementing deep learning methods. These methods not only can light pathologists’ load but also can provide accurate computing results.In this paper, we use U-2-NET as our backbone to perform Binary Image Segmentation on CRC pathology slides. CRC pathology slides have a variety of non-conforming shapes and colors which is an enormous challenge for detecting cancer areas. U-2-NET was originally used in the Salient Object Detection (SOD) task to find the most unique regions of human attention, which can be used to identify abnormal regions in pathological slices. Moreover, the RSU block of U-2-NET can handle long-term and short-term dependencies, which we believe helps maintain contextual information. With the large computational costs, U-2-NET is hard to implement for application. Our purposed method can use preprocessing, image-selecting mechanisms and transfer learning concepts to solve this problem.Our results show that the model trained with a small part of the data set and a modified small object function has the best results for Binary Image Segmentation of colorectal cancer pathology sections by U-2-NET, with the best IOU (0.77) and Dice Loss (0.83) compared with other models (MSRFCNN, FCN, SegNet, and Unet). Furthermore, after transferring learning using pre-trained weights from the SOD dataset, the results are improved compared to those of learning the network from scratch.","PeriodicalId":382440,"journal":{"name":"2022 IEEE International Conference on Recent Advances in Systems Science and Engineering (RASSE)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125618312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-08-29DOI: 10.1109/RASSE54974.2022.9989690
D. Karunakaran, J. S. Berrio, Stewart Worrall, E. Nebot
Autonomous vehicles have the potential to lower the accident rate when compared to human driving. Moreover, it has been the driving force of automated vehicles’ rapid development over the last few years. In the higher Society of Automotive Engineers (SAE) automation level, the vehicle’s and passengers’ safety responsibility is transferred from the driver to the automated system, so thoroughly validating such a system is essential. Recently, academia and industry have embraced scenario-based evaluation as the complementary approach to road testing, reducing the overall testing effort required. It is essential to determine the system’s flaws before deploying it on public roads as there is no safety driver to guarantee the reliability of such a system. This paper proposes a Reinforcement Learning (RL) based scenario-based falsification method to search for a high-risk scenario in a pedestrian crossing traffic situation. We define a scenario as risky when a system under test (SUT) does not satisfy the requirement. The reward function for our RL approach is based on Intel’s Responsibility Sensitive Safety(RSS), Euclidean distance, and distance to a potential collision. Code and videos are available online at https://github.com/dkarunakaran/scenario_based_falsification.
{"title":"Critical Concrete Scenario Generation Using Scenario-Based Falsification","authors":"D. Karunakaran, J. S. Berrio, Stewart Worrall, E. Nebot","doi":"10.1109/RASSE54974.2022.9989690","DOIUrl":"https://doi.org/10.1109/RASSE54974.2022.9989690","url":null,"abstract":"Autonomous vehicles have the potential to lower the accident rate when compared to human driving. Moreover, it has been the driving force of automated vehicles’ rapid development over the last few years. In the higher Society of Automotive Engineers (SAE) automation level, the vehicle’s and passengers’ safety responsibility is transferred from the driver to the automated system, so thoroughly validating such a system is essential. Recently, academia and industry have embraced scenario-based evaluation as the complementary approach to road testing, reducing the overall testing effort required. It is essential to determine the system’s flaws before deploying it on public roads as there is no safety driver to guarantee the reliability of such a system. This paper proposes a Reinforcement Learning (RL) based scenario-based falsification method to search for a high-risk scenario in a pedestrian crossing traffic situation. We define a scenario as risky when a system under test (SUT) does not satisfy the requirement. The reward function for our RL approach is based on Intel’s Responsibility Sensitive Safety(RSS), Euclidean distance, and distance to a potential collision. Code and videos are available online at https://github.com/dkarunakaran/scenario_based_falsification.","PeriodicalId":382440,"journal":{"name":"2022 IEEE International Conference on Recent Advances in Systems Science and Engineering (RASSE)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124536564","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：