Pub Date : 2019-10-31DOI: 10.1109/NSENS49395.2019.9293988
Sheng Chen, Sumei Li, Chengcheng Zhu
SISR (Single image super-resolution) has always been a key problem in image processing field. In recent years, deep learning has been successfully used to SISR reconstruction. However, most of the previous deep learning methods use L2 norm based on pixel pairs as loss function, which results in a high peak signal-to-noise ratio (PSNR) value, but the perception quality has not been improved. When using Generative Adversarial Network (GAN), although it has good perception quality, PSNR is lower. So we’ll generate realistic results when both of them are used well. The image quality evaluation (IQA) network is to evaluate the image quality, so as to obtain good PSNR value and perception quality. In this paper, we use image quality assessment network to guide the SISR reconstruction network. Besides that, our proposed Super-resolution reconstruction of single image method is composed of several our given cross-attention units (CA) and is trained iteratively. Experimental results demonstrate that our method in qualitative and quantitative is better than others.
{"title":"Guided Super-Resolution Restoration of Single Image Based on Image Quality Evaluation Network","authors":"Sheng Chen, Sumei Li, Chengcheng Zhu","doi":"10.1109/NSENS49395.2019.9293988","DOIUrl":"https://doi.org/10.1109/NSENS49395.2019.9293988","url":null,"abstract":"SISR (Single image super-resolution) has always been a key problem in image processing field. In recent years, deep learning has been successfully used to SISR reconstruction. However, most of the previous deep learning methods use L2 norm based on pixel pairs as loss function, which results in a high peak signal-to-noise ratio (PSNR) value, but the perception quality has not been improved. When using Generative Adversarial Network (GAN), although it has good perception quality, PSNR is lower. So we’ll generate realistic results when both of them are used well. The image quality evaluation (IQA) network is to evaluate the image quality, so as to obtain good PSNR value and perception quality. In this paper, we use image quality assessment network to guide the SISR reconstruction network. Besides that, our proposed Super-resolution reconstruction of single image method is composed of several our given cross-attention units (CA) and is trained iteratively. Experimental results demonstrate that our method in qualitative and quantitative is better than others.","PeriodicalId":246485,"journal":{"name":"2019 IEEE THE 2nd INTERNATIONAL CONFERENCE ON MICRO/NANO SENSORS for AI, HEALTHCARE, AND ROBOTICS (NSENS)","volume":"8 12","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120855708","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 : 2019-10-31DOI: 10.1109/NSENS49395.2019.9293954
Yingjiao Rong, Weixuan Ding, Xinan Wang, G. Shi
Today’s science and technology develop rapidly with the passage of time. In the highly competitive environment, everyone’s time is twenty-four hours a day. No one can have more than one minute. Or less have a minute to go, so if you want to improve your competitiveness, efficient scheduling time is very important, people’s daily needs for efficiency are also growing. In the process of moving these vehicles, how to move the same distance with others in less time, this demand and efficiency can also be seen from the tools we now use to produce vehicles (such as tricycles).In the high-precision control system design of the self-balancing vehicle of the tricycle vehicle, in addition to the high-precision inclination sensor to measure the dynamic inclination angle, the steering control has an absolute value encoder to learn the angle of the steering of the tricycle body. More importantly, in addition to the rear axle drive system on the wheel of the tricycle body itself, we also installed a stepper motor to control the balance of the tricycle body, so that the tricycle body can be at a higher speed. Under the steady steering, the tricycle body will not roll over.Dynamic attitude measurement is a very important aspect in the design of high-precision control systems for self-balancing vehicles of tricycle vehicles. Because the motor is an open-loop system, we need to use the tilt sensor for attitude measurement to obtain high-precision angle values, and the obtained angle value and the motor form a closed-loop control system to achieve more precise motor control to control the tricycle body. balance. So we need to measure the angle change very accurately, because a single axial attitude tilt sensor can’t meet our requirements, and because there are too many shortcomings, we can’t get more accurate values in a dynamic environment, so we use A six-axis sensor that helps us get better precision and precision. Finally, we use the VEKF-based algorithm to eliminate the numerical inaccuracy caused by measuring the dynamic tilt angle, and thus obtain calculate the attitude of the self-balancing tricycle and eliminate the errors generated by the sensor. This algorithm can obtain accurate angle values and can be used in a dynamic environment to enable the self-balancing tricycle dynamic vehicle control system to operate stably.
{"title":"Research on Dynamic Attitude Estimation and Control of Tricycle Based on MEMS Sensing","authors":"Yingjiao Rong, Weixuan Ding, Xinan Wang, G. Shi","doi":"10.1109/NSENS49395.2019.9293954","DOIUrl":"https://doi.org/10.1109/NSENS49395.2019.9293954","url":null,"abstract":"Today’s science and technology develop rapidly with the passage of time. In the highly competitive environment, everyone’s time is twenty-four hours a day. No one can have more than one minute. Or less have a minute to go, so if you want to improve your competitiveness, efficient scheduling time is very important, people’s daily needs for efficiency are also growing. In the process of moving these vehicles, how to move the same distance with others in less time, this demand and efficiency can also be seen from the tools we now use to produce vehicles (such as tricycles).In the high-precision control system design of the self-balancing vehicle of the tricycle vehicle, in addition to the high-precision inclination sensor to measure the dynamic inclination angle, the steering control has an absolute value encoder to learn the angle of the steering of the tricycle body. More importantly, in addition to the rear axle drive system on the wheel of the tricycle body itself, we also installed a stepper motor to control the balance of the tricycle body, so that the tricycle body can be at a higher speed. Under the steady steering, the tricycle body will not roll over.Dynamic attitude measurement is a very important aspect in the design of high-precision control systems for self-balancing vehicles of tricycle vehicles. Because the motor is an open-loop system, we need to use the tilt sensor for attitude measurement to obtain high-precision angle values, and the obtained angle value and the motor form a closed-loop control system to achieve more precise motor control to control the tricycle body. balance. So we need to measure the angle change very accurately, because a single axial attitude tilt sensor can’t meet our requirements, and because there are too many shortcomings, we can’t get more accurate values in a dynamic environment, so we use A six-axis sensor that helps us get better precision and precision. Finally, we use the VEKF-based algorithm to eliminate the numerical inaccuracy caused by measuring the dynamic tilt angle, and thus obtain calculate the attitude of the self-balancing tricycle and eliminate the errors generated by the sensor. This algorithm can obtain accurate angle values and can be used in a dynamic environment to enable the self-balancing tricycle dynamic vehicle control system to operate stably.","PeriodicalId":246485,"journal":{"name":"2019 IEEE THE 2nd INTERNATIONAL CONFERENCE ON MICRO/NANO SENSORS for AI, HEALTHCARE, AND ROBOTICS (NSENS)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126513375","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 : 2019-10-31DOI: 10.1109/NSENS49395.2019.9293956
Keqiang Zhu, Zhonghai He, Hui Sun, X. Cai
Biomass is an important parameter in fermentation processes. The estimation of biomass during fermentation usually uses an off-line method, such as optical density at 600 nm (OD600) or the determination of dry cell weight (DCW). Online measurement of biomass via mid-infrared (MIR) spectroscopy has also been published. However, no strict demonstration has been given that biomass measurement by MIR is due to the specific absorption of infrared radiation by cells. Three factors are analyzed about cell absorption, which being: optical sampling theory, spectral absorbance intensity inspection, and PLS regression of cells model. Three aspects lead to the conclusion that the measurement of biomass by MIR is not due to specific absorption by bacteria but rather to a chance correlation with the substrate glutamate in this study. If a chance correlation is present, the biomass can be measured by this indirect method. The most reliable measurement method is still by DCW or OD600. It is frustrating that the online measurement of biomass still remains uncertain.
{"title":"Can biomass be measured in a fermentation process using ATR-FTIR spectroscopy Bacillus subtilis as an example","authors":"Keqiang Zhu, Zhonghai He, Hui Sun, X. Cai","doi":"10.1109/NSENS49395.2019.9293956","DOIUrl":"https://doi.org/10.1109/NSENS49395.2019.9293956","url":null,"abstract":"Biomass is an important parameter in fermentation processes. The estimation of biomass during fermentation usually uses an off-line method, such as optical density at 600 nm (OD600) or the determination of dry cell weight (DCW). Online measurement of biomass via mid-infrared (MIR) spectroscopy has also been published. However, no strict demonstration has been given that biomass measurement by MIR is due to the specific absorption of infrared radiation by cells. Three factors are analyzed about cell absorption, which being: optical sampling theory, spectral absorbance intensity inspection, and PLS regression of cells model. Three aspects lead to the conclusion that the measurement of biomass by MIR is not due to specific absorption by bacteria but rather to a chance correlation with the substrate glutamate in this study. If a chance correlation is present, the biomass can be measured by this indirect method. The most reliable measurement method is still by DCW or OD600. It is frustrating that the online measurement of biomass still remains uncertain.","PeriodicalId":246485,"journal":{"name":"2019 IEEE THE 2nd INTERNATIONAL CONFERENCE ON MICRO/NANO SENSORS for AI, HEALTHCARE, AND ROBOTICS (NSENS)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124159709","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 : 2019-10-31DOI: 10.1109/NSENS49395.2019.9293967
Zhonghai He, Kexin Yang, X. Cai, Hui Sun
Calibration model building is composed of suitable number of samples and multivariate regression techniques. The accuracy of prediction is determined by both factors (steps). In these two steps, the multivariate regression step is influenced by too many factors, making it impossible to determine the number of samples. However, sample number in sample collection step can be used to ensure population representation in statistics. The sample number is the cornerstone of the robustness of model that should be concentrated on; however, few instructions but some empirical expressions have been given up until now. The factors affecting the sampling accuracy include confidence level, relative standard errors, and relative representation requirements. The required number can be calculated by statistical parameters of population and the required representation. The relative standard error is an important factor related to the statistical parameters of the sample set. For general instructions, the calibration kit should use 100-150 samples, the more the better, but it is not recommended to use more than 200. These suggestions would help guide the operator by selecting an appropriate calibration sample number in spectroscopy.
{"title":"Calibration sample number determined by theory of sampling provide threshold for multivariate model building","authors":"Zhonghai He, Kexin Yang, X. Cai, Hui Sun","doi":"10.1109/NSENS49395.2019.9293967","DOIUrl":"https://doi.org/10.1109/NSENS49395.2019.9293967","url":null,"abstract":"Calibration model building is composed of suitable number of samples and multivariate regression techniques. The accuracy of prediction is determined by both factors (steps). In these two steps, the multivariate regression step is influenced by too many factors, making it impossible to determine the number of samples. However, sample number in sample collection step can be used to ensure population representation in statistics. The sample number is the cornerstone of the robustness of model that should be concentrated on; however, few instructions but some empirical expressions have been given up until now. The factors affecting the sampling accuracy include confidence level, relative standard errors, and relative representation requirements. The required number can be calculated by statistical parameters of population and the required representation. The relative standard error is an important factor related to the statistical parameters of the sample set. For general instructions, the calibration kit should use 100-150 samples, the more the better, but it is not recommended to use more than 200. These suggestions would help guide the operator by selecting an appropriate calibration sample number in spectroscopy.","PeriodicalId":246485,"journal":{"name":"2019 IEEE THE 2nd INTERNATIONAL CONFERENCE ON MICRO/NANO SENSORS for AI, HEALTHCARE, AND ROBOTICS (NSENS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124656836","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 : 2019-10-31DOI: 10.1109/NSENS49395.2019.9293963
Shengli Zhou, Kuiying Yin
The current sensing devices for measuring continuous finger movement are either restrictive to users (data glove) or easily influenced by external environment (optical or magnetic trackers based method). Therefore, the objective of this study is developing a continuous finger movement tracking system that is more easy and comfortable to use. The surface electromyography (sEMG) signals applied in this study were collected from human forearm with 10 electrodes, and transmitted to the computer via cables. Timedomain features were extracted and further filtered with a low-pass filter to smooth the features. Three partial least square regression (PLSR) based movement estimation models had been built for the three movements investigated in this study, and one movement recognition model was constructed to determine which movement estimation model would be applied for the new incoming samples. The prediction accuracy evaluated in terms of Pearson’s correlation coefficient ranges from 0.84 to 0.91 for single finger flexion, and ranges from 0.53 to 0.83 for the movement of fingers flexed together in fist. The normalized root mean square error (NRMSE) ranges from 0.04 to 0.1 for single finger flexion, and ranges from 0.046 to 0.14 for the movement of fingers flexed together in fist. The effectiveness of PLSR has also been proved by comparing its performance with linear regression (LR) model.
{"title":"Continuous Finger Joint Angle Estimation With sEMG Signals","authors":"Shengli Zhou, Kuiying Yin","doi":"10.1109/NSENS49395.2019.9293963","DOIUrl":"https://doi.org/10.1109/NSENS49395.2019.9293963","url":null,"abstract":"The current sensing devices for measuring continuous finger movement are either restrictive to users (data glove) or easily influenced by external environment (optical or magnetic trackers based method). Therefore, the objective of this study is developing a continuous finger movement tracking system that is more easy and comfortable to use. The surface electromyography (sEMG) signals applied in this study were collected from human forearm with 10 electrodes, and transmitted to the computer via cables. Timedomain features were extracted and further filtered with a low-pass filter to smooth the features. Three partial least square regression (PLSR) based movement estimation models had been built for the three movements investigated in this study, and one movement recognition model was constructed to determine which movement estimation model would be applied for the new incoming samples. The prediction accuracy evaluated in terms of Pearson’s correlation coefficient ranges from 0.84 to 0.91 for single finger flexion, and ranges from 0.53 to 0.83 for the movement of fingers flexed together in fist. The normalized root mean square error (NRMSE) ranges from 0.04 to 0.1 for single finger flexion, and ranges from 0.046 to 0.14 for the movement of fingers flexed together in fist. The effectiveness of PLSR has also been proved by comparing its performance with linear regression (LR) model.","PeriodicalId":246485,"journal":{"name":"2019 IEEE THE 2nd INTERNATIONAL CONFERENCE ON MICRO/NANO SENSORS for AI, HEALTHCARE, AND ROBOTICS (NSENS)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121603477","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 : 2019-10-31DOI: 10.1109/NSENS49395.2019.9293994
Fei Fei, Ying Leng, Min Yang, Changcheng Wu, Dehua Yang
Foot plantar pressure provides plenty of information for gait research and medical diagnostics. Gait analysis can be used to evaluate stroke patient’s mobility and rehabilitation status. However, most of existing gait analysis system can only be used in laboratory or indoor occasions. It makes a large limitation for the gait data collection and analysis. This paper presents a novel wearable human gait analysis system based on flexible circuit and piezoresistive pressure sensors. The insole embedded with 8 pressure sensors is fabricated to collect dynamic resistance varying signals due to the piezoresistive effect. Then the resistance signal is converted to voltage signals with a resistance-voltage conversion circuit board. The wireless transmitter sends the gait data to computer for real-time gait analysis via WIFI chip. The experiment results show the pressure difference on different area of foot plantar during walking, running and squatting. And several gait characteristics such as peak-peak voltage and mean voltage are also calculated and compared. It shows that this novel wearable insole device can be used to monitor plantar pressure during daily life effectively.
{"title":"Development of A Wearable Human Gait Analysis System Based on Plantar Pressure Sensors","authors":"Fei Fei, Ying Leng, Min Yang, Changcheng Wu, Dehua Yang","doi":"10.1109/NSENS49395.2019.9293994","DOIUrl":"https://doi.org/10.1109/NSENS49395.2019.9293994","url":null,"abstract":"Foot plantar pressure provides plenty of information for gait research and medical diagnostics. Gait analysis can be used to evaluate stroke patient’s mobility and rehabilitation status. However, most of existing gait analysis system can only be used in laboratory or indoor occasions. It makes a large limitation for the gait data collection and analysis. This paper presents a novel wearable human gait analysis system based on flexible circuit and piezoresistive pressure sensors. The insole embedded with 8 pressure sensors is fabricated to collect dynamic resistance varying signals due to the piezoresistive effect. Then the resistance signal is converted to voltage signals with a resistance-voltage conversion circuit board. The wireless transmitter sends the gait data to computer for real-time gait analysis via WIFI chip. The experiment results show the pressure difference on different area of foot plantar during walking, running and squatting. And several gait characteristics such as peak-peak voltage and mean voltage are also calculated and compared. It shows that this novel wearable insole device can be used to monitor plantar pressure during daily life effectively.","PeriodicalId":246485,"journal":{"name":"2019 IEEE THE 2nd INTERNATIONAL CONFERENCE ON MICRO/NANO SENSORS for AI, HEALTHCARE, AND ROBOTICS (NSENS)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127706210","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 : 2019-10-31DOI: 10.1109/NSENS49395.2019.9293996
Tejal Gala, Yanwen Xiong, Min Hubbard, Winn Hong, J. Mai
Patient compliance during drug trials and adherence to treatment regimens after a medical diagnosis are known pervasive problems in the practice of medicine. Any practical solution to this problem will require an easy method to identify and to verify the administration of orally-ingested drugs. Deep learning algorithms were applied to images of drugs in pill form. These images were taken using both a smart phone camera and using a hyperspectral imager based on a low-cost CMOS camera. As a proof-of-concept demonstration, 1, 7SS images were taken using a normal CMOS camera of four common pill types. The images of acetaminophen, acetylsalicylic acid and ibuprofen were taken using various backgrounds, image angles, and lighting conditions. The results show over 90% accuracy when the convolutional neural network is trained and tested using only normal camera images. The results improved to 100% when trained and tested using4 baseline “datacubes” taken with a low-cost hyperspectral camera solution; however, due to matrix dimensional differences, a ID CNN was used in this case, while a 2D CNN was used with the normal camera images. Each hyperspectral cube included information from effectively 31 wavebands. With more hyperspectral images to expand the drug training set, this approach would be promising for daily use to quickly identify similar pills in the clinical or home environment as well as in smart phone apps to remotely monitor patient compliance to a drug-based treatment regimen.
{"title":"Deep Learning with Hyperspectral and Normal Camera Images for Automated Recognition of Orally-administered Drugs","authors":"Tejal Gala, Yanwen Xiong, Min Hubbard, Winn Hong, J. Mai","doi":"10.1109/NSENS49395.2019.9293996","DOIUrl":"https://doi.org/10.1109/NSENS49395.2019.9293996","url":null,"abstract":"Patient compliance during drug trials and adherence to treatment regimens after a medical diagnosis are known pervasive problems in the practice of medicine. Any practical solution to this problem will require an easy method to identify and to verify the administration of orally-ingested drugs. Deep learning algorithms were applied to images of drugs in pill form. These images were taken using both a smart phone camera and using a hyperspectral imager based on a low-cost CMOS camera. As a proof-of-concept demonstration, 1, 7SS images were taken using a normal CMOS camera of four common pill types. The images of acetaminophen, acetylsalicylic acid and ibuprofen were taken using various backgrounds, image angles, and lighting conditions. The results show over 90% accuracy when the convolutional neural network is trained and tested using only normal camera images. The results improved to 100% when trained and tested using4 baseline “datacubes” taken with a low-cost hyperspectral camera solution; however, due to matrix dimensional differences, a ID CNN was used in this case, while a 2D CNN was used with the normal camera images. Each hyperspectral cube included information from effectively 31 wavebands. With more hyperspectral images to expand the drug training set, this approach would be promising for daily use to quickly identify similar pills in the clinical or home environment as well as in smart phone apps to remotely monitor patient compliance to a drug-based treatment regimen.","PeriodicalId":246485,"journal":{"name":"2019 IEEE THE 2nd INTERNATIONAL CONFERENCE ON MICRO/NANO SENSORS for AI, HEALTHCARE, AND ROBOTICS (NSENS)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114605611","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 : 2019-10-31DOI: 10.1109/NSENS49395.2019.9293955
Pan Li, Haibo Yu, Feifei Wang, Gwo-Bin Lee, Lianqing Liu, W. Li
Nanoscale material surface patterning on semiconductor requires multi-scale measurements for the determination of their geometric dimensions and shapes. In this paper, we propose an in-situ,real-time super-resolution imaging technique using a simple microsphere microlens to monitor the dynamic process of photo-assisted electrochemical printing in a microfluidic chip. The microsphere microlens with diameters of 30 ~ 60 $mu$ m were set close to a semiconductor surface to image the electrochemical printing process underneath. With the microsphere-based imaging system, both the depositing process of silver nanoparticles with 200 nm ~ 300 nm in diameter and the growing process of silver belts were observed. Also, we experimentally observed how a typical 120° angle formed at the terminal of a silver belt through the microsphere superlens. Super-resolution monitoring ability provided by microsphere lens will shine a light on the sub-diffraction process of micro/nano fabrication.
{"title":"Super-resolution Monitoring of React-on-demand Photo-assisted Electrochemical Printing via Microsphere Nanoscopy","authors":"Pan Li, Haibo Yu, Feifei Wang, Gwo-Bin Lee, Lianqing Liu, W. Li","doi":"10.1109/NSENS49395.2019.9293955","DOIUrl":"https://doi.org/10.1109/NSENS49395.2019.9293955","url":null,"abstract":"Nanoscale material surface patterning on semiconductor requires multi-scale measurements for the determination of their geometric dimensions and shapes. In this paper, we propose an in-situ,real-time super-resolution imaging technique using a simple microsphere microlens to monitor the dynamic process of photo-assisted electrochemical printing in a microfluidic chip. The microsphere microlens with diameters of 30 ~ 60 $mu$ m were set close to a semiconductor surface to image the electrochemical printing process underneath. With the microsphere-based imaging system, both the depositing process of silver nanoparticles with 200 nm ~ 300 nm in diameter and the growing process of silver belts were observed. Also, we experimentally observed how a typical 120° angle formed at the terminal of a silver belt through the microsphere superlens. Super-resolution monitoring ability provided by microsphere lens will shine a light on the sub-diffraction process of micro/nano fabrication.","PeriodicalId":246485,"journal":{"name":"2019 IEEE THE 2nd INTERNATIONAL CONFERENCE ON MICRO/NANO SENSORS for AI, HEALTHCARE, AND ROBOTICS (NSENS)","volume":"241 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121153664","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 : 2019-10-31DOI: 10.1109/NSENS49395.2019.9293953
Sheng Chen, Sumei Li, Chengcheng Zhu
The Super-Resolution Generative Adversarial Network (SRGAN) is a seminal work that is capable of generating more realistic texture in semantics and style during single image super-resolution. However, Since the loss function adopts L2 norm based on pixel points, the hallucinated details are often accompanied with unpleasant artifacts even false pixels. Our model adjusts generative loss to L1 norm, and perceptual loss is still based on L2 norm. L1 cost function can reduce the coefficients of some features to zero, thus indirectly realizing the selection of features according to the perceptual loss, and obtaining more real texture features. The combination of these two loss functions ensures that the reconstructed results of the model are very close to the target image in terms of spatial features, high-level abstract features and semantic features, overall sensory and image quality. The generating network of our model is based on dense residual structure, and the dense connection of residual-in-residual is used to implement fast and accurate learning of high frequency features of images. The adversarial network is based on the structure of discriminators in DCGAN and WGAN. Experimental results show that subjective quality we reconstructed is much higher than SRGAN.
{"title":"Advanced Generative Adversarial Network Based on Dense Connection For Single Image Super Resolution","authors":"Sheng Chen, Sumei Li, Chengcheng Zhu","doi":"10.1109/NSENS49395.2019.9293953","DOIUrl":"https://doi.org/10.1109/NSENS49395.2019.9293953","url":null,"abstract":"The Super-Resolution Generative Adversarial Network (SRGAN) is a seminal work that is capable of generating more realistic texture in semantics and style during single image super-resolution. However, Since the loss function adopts L2 norm based on pixel points, the hallucinated details are often accompanied with unpleasant artifacts even false pixels. Our model adjusts generative loss to L1 norm, and perceptual loss is still based on L2 norm. L1 cost function can reduce the coefficients of some features to zero, thus indirectly realizing the selection of features according to the perceptual loss, and obtaining more real texture features. The combination of these two loss functions ensures that the reconstructed results of the model are very close to the target image in terms of spatial features, high-level abstract features and semantic features, overall sensory and image quality. The generating network of our model is based on dense residual structure, and the dense connection of residual-in-residual is used to implement fast and accurate learning of high frequency features of images. The adversarial network is based on the structure of discriminators in DCGAN and WGAN. Experimental results show that subjective quality we reconstructed is much higher than SRGAN.","PeriodicalId":246485,"journal":{"name":"2019 IEEE THE 2nd INTERNATIONAL CONFERENCE ON MICRO/NANO SENSORS for AI, HEALTHCARE, AND ROBOTICS (NSENS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131959640","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 : 2019-10-31DOI: 10.1109/NSENS49395.2019.9293999
Kai Wang, Kingshing Yip, Chengchun Shien, Xinan Wang, G. Shi
The rapid development of micro/nano-technology in recent years, an inertial navigation system (INS) composed of MEMS devices is widely used in various fields such as detection, machinery, transportation, and military affairs. The trend of using MEMS devices for navigation in the field of transportation is increasing. The three-dimensional MEMS electronic compass mainly includes a magnetometer and accelerometer. It mainly uses the earth’s magnetic field, gravity acceleration and other parameters to provide the bearing and attitude of the carrier for the navigation system. However, the current inertial navigation system is easy to get lost when it encounters magnetic disturbance, and the irregular movement process is easy to cause errors, even in the static environment is not accurate. In order to solve this problem, in this paper, a dynamic heading filtering algorithm based on Extended Kalman Filter is proposed. The two sets of sensors are used for the positive phase installation, and an extended Kalman filter algorithm is designed. The heading solution can be self-adaption according to different magnetic disturbances. Finally, the experiment verified that the heading error of the carrier after filtering into the dynamic condition was ± 1°, which has met the actual requirements.
{"title":"Research on Dynamic Heading Calculation of Complex Magnetic Disturbance","authors":"Kai Wang, Kingshing Yip, Chengchun Shien, Xinan Wang, G. Shi","doi":"10.1109/NSENS49395.2019.9293999","DOIUrl":"https://doi.org/10.1109/NSENS49395.2019.9293999","url":null,"abstract":"The rapid development of micro/nano-technology in recent years, an inertial navigation system (INS) composed of MEMS devices is widely used in various fields such as detection, machinery, transportation, and military affairs. The trend of using MEMS devices for navigation in the field of transportation is increasing. The three-dimensional MEMS electronic compass mainly includes a magnetometer and accelerometer. It mainly uses the earth’s magnetic field, gravity acceleration and other parameters to provide the bearing and attitude of the carrier for the navigation system. However, the current inertial navigation system is easy to get lost when it encounters magnetic disturbance, and the irregular movement process is easy to cause errors, even in the static environment is not accurate. In order to solve this problem, in this paper, a dynamic heading filtering algorithm based on Extended Kalman Filter is proposed. The two sets of sensors are used for the positive phase installation, and an extended Kalman filter algorithm is designed. The heading solution can be self-adaption according to different magnetic disturbances. Finally, the experiment verified that the heading error of the carrier after filtering into the dynamic condition was ± 1°, which has met the actual requirements.","PeriodicalId":246485,"journal":{"name":"2019 IEEE THE 2nd INTERNATIONAL CONFERENCE ON MICRO/NANO SENSORS for AI, HEALTHCARE, AND ROBOTICS (NSENS)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121500290","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：