Yuqing Zhao, Guangyuan Fu, Hongqiao Wang, Shaolei Zhang, Min Yue
Blind deblurring of a single infrared image is a challenging computer vision problem. Because the blur is not only caused by the motion of different objects but also by the relative motion and jitter of cameras, there is a change of scene depth. In this work, a method based on the GAN and channel prior discrimination is proposed for infrared image deblurring. Different from the previous work, we combine the traditional blind deblurring method and the blind deblurring method based on the learning method, and uniform and nonuniform blurred images are considered, respectively. By training the proposed model on different datasets, it is proved that the proposed method achieves competitive performance in terms of deblurring quality (objective and subjective).
{"title":"Infrared Image Deblurring Based on Generative Adversarial Networks","authors":"Yuqing Zhao, Guangyuan Fu, Hongqiao Wang, Shaolei Zhang, Min Yue","doi":"10.1155/2021/9946809","DOIUrl":"https://doi.org/10.1155/2021/9946809","url":null,"abstract":"Blind deblurring of a single infrared image is a challenging computer vision problem. Because the blur is not only caused by the motion of different objects but also by the relative motion and jitter of cameras, there is a change of scene depth. In this work, a method based on the GAN and channel prior discrimination is proposed for infrared image deblurring. Different from the previous work, we combine the traditional blind deblurring method and the blind deblurring method based on the learning method, and uniform and nonuniform blurred images are considered, respectively. By training the proposed model on different datasets, it is proved that the proposed method achieves competitive performance in terms of deblurring quality (objective and subjective).","PeriodicalId":55995,"journal":{"name":"International Journal of Optics","volume":" ","pages":"1-16"},"PeriodicalIF":1.7,"publicationDate":"2021-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49035079","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Optical fibers with high sensitivity are in demand due to their great potential in sensor application. Semiconductors, such as ZnO, are good materials. Using them as a second cladding offers opportunities in realizing next-generation multimaterial fiber optics. COMSOL Multiphysics is used to simulate heterocore macrobend fiber optics with the same curvature radius but different values of refractive index and thickness of ZnO films. The optimum thickness of ZnO films is identified by determining the loss of optical fibers. Macrobend heterocore fiber optics by adding ZnO thin film has been established by simulating and interpreting the relationship in terms of transmission and refractive index in the evanescent field. These results will provide a reliable fundamental to guide the performance in practice.
{"title":"Sensitivity Enhancement of Heterocore Macrobend Fiber Optics by Adding a ZnO Film","authors":"N. A. M. Mohd Arif, D. Berhanuddin, A. Ehsan","doi":"10.1155/2021/5511342","DOIUrl":"https://doi.org/10.1155/2021/5511342","url":null,"abstract":"Optical fibers with high sensitivity are in demand due to their great potential in sensor application. Semiconductors, such as ZnO, are good materials. Using them as a second cladding offers opportunities in realizing next-generation multimaterial fiber optics. COMSOL Multiphysics is used to simulate heterocore macrobend fiber optics with the same curvature radius but different values of refractive index and thickness of ZnO films. The optimum thickness of ZnO films is identified by determining the loss of optical fibers. Macrobend heterocore fiber optics by adding ZnO thin film has been established by simulating and interpreting the relationship in terms of transmission and refractive index in the evanescent field. These results will provide a reliable fundamental to guide the performance in practice.","PeriodicalId":55995,"journal":{"name":"International Journal of Optics","volume":"2021 1","pages":"1-8"},"PeriodicalIF":1.7,"publicationDate":"2021-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43435998","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Leijun Hu, Liwen Sheng, Jisong Yan, Ligong Li, M. Yuan, Fu Sun, F. Nian, Long Li, Jiaqing Liu, Shuai Zhou, Zhiming Liu
A multiparameter Brillouin fiber-optic sensor for distributed strain and temperature information measuring based on spontaneous scattering in a common communication optical fiber (the G. 652. D commercial fiber) is presented and experimentally demonstrated. Benefiting from the difference of the temperature and strain sensitivity from different Brillouin peaks with different acoustic modes, our proposed sensing configuration can be used to distinguish ambient temperature and applied strain at the same time, which is an excellent candidate to address the problem of cross-sensitivity in the classical Brillouin system. In the experimental section, using a 21.8 km sensing length of communication optical fiber, a temperature accuracy of 1.13°C and a strain accuracy of 21.46 μe are obtained simultaneously. Considering the performance we achieved now, the proposed innovation and experimental setup will have some potential applications in the field of fiber sensors.
{"title":"Simultaneous Measurement of Distributed Temperature and Strain through Brillouin Frequency Shift Using a Common Communication Optical Fiber","authors":"Leijun Hu, Liwen Sheng, Jisong Yan, Ligong Li, M. Yuan, Fu Sun, F. Nian, Long Li, Jiaqing Liu, Shuai Zhou, Zhiming Liu","doi":"10.1155/2021/6610674","DOIUrl":"https://doi.org/10.1155/2021/6610674","url":null,"abstract":"A multiparameter Brillouin fiber-optic sensor for distributed strain and temperature information measuring based on spontaneous scattering in a common communication optical fiber (the G. 652. D commercial fiber) is presented and experimentally demonstrated. Benefiting from the difference of the temperature and strain sensitivity from different Brillouin peaks with different acoustic modes, our proposed sensing configuration can be used to distinguish ambient temperature and applied strain at the same time, which is an excellent candidate to address the problem of cross-sensitivity in the classical Brillouin system. In the experimental section, using a 21.8 km sensing length of communication optical fiber, a temperature accuracy of 1.13°C and a strain accuracy of 21.46 μe are obtained simultaneously. Considering the performance we achieved now, the proposed innovation and experimental setup will have some potential applications in the field of fiber sensors.","PeriodicalId":55995,"journal":{"name":"International Journal of Optics","volume":"2021 1","pages":"1-6"},"PeriodicalIF":1.7,"publicationDate":"2021-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42152081","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Liwen Sheng, Jisong Yan, Ligong Li, M. Yuan, Shuai Zhou, Ruilin Xu, Jiaqing Liu, F. Nian, Long Li, Zhiming Liu
Utilizing a single-photon detector, a novel direct-detection optical-fiber sensor for distributed measurement of temperature based on spontaneous Brillouin scattering is proposed and demonstrated experimentally. In our scheme, the ratio of the backscattered Rayleigh signal and the backscattered Brillouin anti-Stokes is adopted to retrieve the monitored temperature information along the optical fiber. Taking advantage of the high sensitivity of the single-photon detector, our proposed system achieves a dynamic range of 20 dB without any optical amplification. The obtainable dynamic range corresponds to a sensing distance of 120 km with a measured temperature error of 0.96°C. Furthermore, the proof-of-concept experiment demonstrates 1.2 m spatial resolution over 4.2 km sensing link with 1.24°C temperature error. Considering the performance we achieved now, and the increasing improvement of the fabrication technology of sing-photon detector, the photon-counting distributed Brillouin sensor is opening a door in the field of optical-fiber sensors.
{"title":"Distributed Temperature Sensing System Based on Brillouin Scattering Effect Using a Single-Photon Detector","authors":"Liwen Sheng, Jisong Yan, Ligong Li, M. Yuan, Shuai Zhou, Ruilin Xu, Jiaqing Liu, F. Nian, Long Li, Zhiming Liu","doi":"10.1155/2021/6623987","DOIUrl":"https://doi.org/10.1155/2021/6623987","url":null,"abstract":"Utilizing a single-photon detector, a novel direct-detection optical-fiber sensor for distributed measurement of temperature based on spontaneous Brillouin scattering is proposed and demonstrated experimentally. In our scheme, the ratio of the backscattered Rayleigh signal and the backscattered Brillouin anti-Stokes is adopted to retrieve the monitored temperature information along the optical fiber. Taking advantage of the high sensitivity of the single-photon detector, our proposed system achieves a dynamic range of 20 dB without any optical amplification. The obtainable dynamic range corresponds to a sensing distance of 120 km with a measured temperature error of 0.96°C. Furthermore, the proof-of-concept experiment demonstrates 1.2 m spatial resolution over 4.2 km sensing link with 1.24°C temperature error. Considering the performance we achieved now, and the increasing improvement of the fabrication technology of sing-photon detector, the photon-counting distributed Brillouin sensor is opening a door in the field of optical-fiber sensors.","PeriodicalId":55995,"journal":{"name":"International Journal of Optics","volume":"2021 1","pages":"1-9"},"PeriodicalIF":1.7,"publicationDate":"2021-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41485474","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shan Wei, Yajun Pang, Zhen-xu Bai, Yulei Wang, Zhiwei Lu
It is of great significance to measure the residual stress distribution accurately for optical elements and evaluate its influence on the performance of optical instruments in optical imaging, aviation remote sensing, semiconductor manufacturing, and other fields. The stress of optical elements can be closely related to birefringence based on photoelasticity. Thus, the method of quantifying birefringence to obtain the stress becomes the main method of stress measurement technologies for optical elements. This paper first introduces the basic principle of stress measurement based on photoelasticity. Then, the research progress of stress measurement technologies based on this principle is reviewed, which can be classified into two methods: polarization method and interference method. Meanwhile, the advantages and disadvantages of various stress measurement technologies are analyzed and compared. Finally, the developing trend of stress measurement technologies for optical elements is summarized and prospected.
{"title":"Research Progress of Stress Measurement Technologies for Optical Elements","authors":"Shan Wei, Yajun Pang, Zhen-xu Bai, Yulei Wang, Zhiwei Lu","doi":"10.1155/2021/5541358","DOIUrl":"https://doi.org/10.1155/2021/5541358","url":null,"abstract":"It is of great significance to measure the residual stress distribution accurately for optical elements and evaluate its influence on the performance of optical instruments in optical imaging, aviation remote sensing, semiconductor manufacturing, and other fields. The stress of optical elements can be closely related to birefringence based on photoelasticity. Thus, the method of quantifying birefringence to obtain the stress becomes the main method of stress measurement technologies for optical elements. This paper first introduces the basic principle of stress measurement based on photoelasticity. Then, the research progress of stress measurement technologies based on this principle is reviewed, which can be classified into two methods: polarization method and interference method. Meanwhile, the advantages and disadvantages of various stress measurement technologies are analyzed and compared. Finally, the developing trend of stress measurement technologies for optical elements is summarized and prospected.","PeriodicalId":55995,"journal":{"name":"International Journal of Optics","volume":"2021 1","pages":"1-11"},"PeriodicalIF":1.7,"publicationDate":"2021-04-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42739725","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A new asymmetric optical double image encryption algorithm is proposed, which combines phase truncation and singular value decomposition. *e plain text is encrypted with two-stage phase keys to obtain a uniformly distributed cipher text and two new decryption keys. *ese keys are generated during the encryption process and are different from encryption keys. It realizes asymmetric encryption and improves the security of the system. *e unscrambling keys in the encryption operation are mainly related to plain text. At the same time, the system is more resistant to selective plain text attacks; it also improves the sensitivity of decryption keys. With the application of phase truncation, the key space expanded and the security of the cryptographic system is enhanced. *e efficacy of the system is calculated by evaluating the estimated error between the input and retrieved images. *e proposed technique provides innumerable security keys and is robust against various potential attacks. Numerical simulations verify the effectiveness and security of the proposed technique.
{"title":"A Secure Asymmetric Optical Image Encryption Based on Phase Truncation and Singular Value Decomposition in Linear Canonical Transform Domain","authors":"Anshula Sangwan, Hukum Singh","doi":"10.1155/2021/5510125","DOIUrl":"https://doi.org/10.1155/2021/5510125","url":null,"abstract":"A new asymmetric optical double image encryption algorithm is proposed, which combines phase truncation and singular value decomposition. *e plain text is encrypted with two-stage phase keys to obtain a uniformly distributed cipher text and two new decryption keys. *ese keys are generated during the encryption process and are different from encryption keys. It realizes asymmetric encryption and improves the security of the system. *e unscrambling keys in the encryption operation are mainly related to plain text. At the same time, the system is more resistant to selective plain text attacks; it also improves the sensitivity of decryption keys. With the application of phase truncation, the key space expanded and the security of the cryptographic system is enhanced. *e efficacy of the system is calculated by evaluating the estimated error between the input and retrieved images. *e proposed technique provides innumerable security keys and is robust against various potential attacks. Numerical simulations verify the effectiveness and security of the proposed technique.","PeriodicalId":55995,"journal":{"name":"International Journal of Optics","volume":"2021 1","pages":"1-19"},"PeriodicalIF":1.7,"publicationDate":"2021-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45337214","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Frequency-modulated continuous-wave (FMCW) interference, as a new technology of laser interferometry, has the advantages of length traceability, large range, high accuracy, simple structure, and optical fiber transmission. Based on the formula of FMCW laser interference displacement, a zero-crossing phase detection algorithm is proposed, which can accurately calculate the initial phase of a cosine signal in a modulation period, and it is successfully applied to the contact laser interference displacement sensor. The experimental results show that the FMCW technology based on the zero-crossing phase detection algorithm can achieve the technical specifications of the contact displacement sensor with a measurement range greater than 15 mm and the standard deviation is less than 0.01 μm. The conversion of noncontact measurement to contact measurement can realize the direct measurement of workpieces with complex surface conditions on the production line, breaking through the limitation of optical measurement and expanding the application of optical fiber interferometry.
{"title":"Application Research of Frequency-Modulated Continuous-Wave Displacement Sensor Based on Zero-Crossing Phase Detecting Algorithm","authors":"Bin Sun, Junfang Song, G. Zheng, Xiongxing Zhang","doi":"10.1155/2021/6669539","DOIUrl":"https://doi.org/10.1155/2021/6669539","url":null,"abstract":"Frequency-modulated continuous-wave (FMCW) interference, as a new technology of laser interferometry, has the advantages of length traceability, large range, high accuracy, simple structure, and optical fiber transmission. Based on the formula of FMCW laser interference displacement, a zero-crossing phase detection algorithm is proposed, which can accurately calculate the initial phase of a cosine signal in a modulation period, and it is successfully applied to the contact laser interference displacement sensor. The experimental results show that the FMCW technology based on the zero-crossing phase detection algorithm can achieve the technical specifications of the contact displacement sensor with a measurement range greater than 15 mm and the standard deviation is less than 0.01 μm. The conversion of noncontact measurement to contact measurement can realize the direct measurement of workpieces with complex surface conditions on the production line, breaking through the limitation of optical measurement and expanding the application of optical fiber interferometry.","PeriodicalId":55995,"journal":{"name":"International Journal of Optics","volume":" ","pages":""},"PeriodicalIF":1.7,"publicationDate":"2021-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44522811","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Corrigendum to “Phase Singularities to Polarization Singularities”","authors":"Ruchi, P. Senthilkumaran, S. Pal","doi":"10.1155/2021/1075165","DOIUrl":"https://doi.org/10.1155/2021/1075165","url":null,"abstract":"<jats:p />","PeriodicalId":55995,"journal":{"name":"International Journal of Optics","volume":"2021 1","pages":"1-1"},"PeriodicalIF":1.7,"publicationDate":"2021-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49594598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We demonstrate an efficient scheme to accelerate the self-similar pulse evolution and reduce the intensity noise of a free-running femtosecond fiber amplifier based on the pump wavelength optimization. Experiments and simulations indicate the enhanced tolerances of the pulse self-similar amplification to the seed signal power and pump wavelength fluctuations, with the optimum 915 nm pump wavelength. ∼20% increase in the compressed pulse quality and ∼31% reduction in the amplifier root-mean-square (RMS) relative intensity noise (RIN) (1.5 kHz to 5 MHz) have been observed, even with a more than 4 times higher pump laser diode (LD) RIN than the case of 976 nm. ∼50 fs transform-limited pulses are generated with the ∼0.03% amplifier RMS RIN. The proposed scheme can lower the requirements of low-noise self-similar femtosecond fiber amplifiers on the power stability of the seed oscillator and the thermal control of the pump LD, thus denoting potentials for the various satellite-based high-precision applications of femtosecond laser in space.
{"title":"Performance Enhancements of Femtosecond Fiber Amplifier by Pump Wavelength Optimization","authors":"Sijia Wang, Peng Qin, Huibin Wang","doi":"10.1155/2021/5514089","DOIUrl":"https://doi.org/10.1155/2021/5514089","url":null,"abstract":"We demonstrate an efficient scheme to accelerate the self-similar pulse evolution and reduce the intensity noise of a free-running femtosecond fiber amplifier based on the pump wavelength optimization. Experiments and simulations indicate the enhanced tolerances of the pulse self-similar amplification to the seed signal power and pump wavelength fluctuations, with the optimum 915 nm pump wavelength. ∼20% increase in the compressed pulse quality and ∼31% reduction in the amplifier root-mean-square (RMS) relative intensity noise (RIN) (1.5 kHz to 5 MHz) have been observed, even with a more than 4 times higher pump laser diode (LD) RIN than the case of 976 nm. ∼50 fs transform-limited pulses are generated with the ∼0.03% amplifier RMS RIN. The proposed scheme can lower the requirements of low-noise self-similar femtosecond fiber amplifiers on the power stability of the seed oscillator and the thermal control of the pump LD, thus denoting potentials for the various satellite-based high-precision applications of femtosecond laser in space.","PeriodicalId":55995,"journal":{"name":"International Journal of Optics","volume":" ","pages":""},"PeriodicalIF":1.7,"publicationDate":"2021-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45920807","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A photonic artificial intelligence chip is based on an optical neural network (ONN), low power consumption, low delay, and strong antiinterference ability. The all-optical diffractive deep neural network has recently demonstrated its inference capabilities on the image classification task. However, the size of the physical model does not have miniaturization and integration, and the optical nonlinearity is not incorporated into the diffraction neural network. By introducing the nonlinear characteristics of the network, complex tasks can be completed with high accuracy. In this study, a nonlinear all-optical diffraction deep neural network (N-D2NN) model based on 10.6 μm wavelength is constructed by combining the ONN and complex-valued neural networks with the nonlinear activation function introduced into the structure. To be specific, the improved activation function of the rectified linear unit (ReLU), i.e., Leaky-ReLU, parametric ReLU (PReLU), and randomized ReLU (RReLU), is selected as the activation function of the N-D2NN model. Through numerical simulation, it is proved that the N-D2NN model based on 10.6 μm wavelength has excellent representation ability, which enables them to perform classification learning tasks of the MNIST handwritten digital dataset and Fashion-MNIST dataset well, respectively. The results show that the N-D2NN model with the RReLU activation function has the highest classification accuracy of 97.86% and 89.28%, respectively. These results provide a theoretical basis for the preparation of miniaturized and integrated N-D2NN model photonic artificial intelligence chips.
{"title":"Nonlinear All-Optical Diffractive Deep Neural Network with 10.6 μm Wavelength for Image Classification","authors":"Yichen Sun, M. Dong, Mingxin Yu, Jiabin Xia, Xu Zhang, Yuchen Bai, Lidan Lu, Lianqing Zhu","doi":"10.1155/2021/6667495","DOIUrl":"https://doi.org/10.1155/2021/6667495","url":null,"abstract":"A photonic artificial intelligence chip is based on an optical neural network (ONN), low power consumption, low delay, and strong antiinterference ability. The all-optical diffractive deep neural network has recently demonstrated its inference capabilities on the image classification task. However, the size of the physical model does not have miniaturization and integration, and the optical nonlinearity is not incorporated into the diffraction neural network. By introducing the nonlinear characteristics of the network, complex tasks can be completed with high accuracy. In this study, a nonlinear all-optical diffraction deep neural network (N-D2NN) model based on 10.6 μm wavelength is constructed by combining the ONN and complex-valued neural networks with the nonlinear activation function introduced into the structure. To be specific, the improved activation function of the rectified linear unit (ReLU), i.e., Leaky-ReLU, parametric ReLU (PReLU), and randomized ReLU (RReLU), is selected as the activation function of the N-D2NN model. Through numerical simulation, it is proved that the N-D2NN model based on 10.6 μm wavelength has excellent representation ability, which enables them to perform classification learning tasks of the MNIST handwritten digital dataset and Fashion-MNIST dataset well, respectively. The results show that the N-D2NN model with the RReLU activation function has the highest classification accuracy of 97.86% and 89.28%, respectively. These results provide a theoretical basis for the preparation of miniaturized and integrated N-D2NN model photonic artificial intelligence chips.","PeriodicalId":55995,"journal":{"name":"International Journal of Optics","volume":" ","pages":"1-16"},"PeriodicalIF":1.7,"publicationDate":"2021-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49603486","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","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：