Pub Date : 2023-11-01DOI: 10.1117/1.OE.62.11.118105
Yang Liu, Fulong Yi, Yuhua Ma, Yongfu Wang, Dianhui Wang
Abstract. Compared with visible light imaging technology, infrared optical imaging technology (see Fig. 1) is less affected by weather and illumination and is a potential auxiliary solution method for the future of the autonomous driving system. Based on the above characteristics, this study proposes the hold surrounding’s key (HSK)-you only look once (YOLOv7) algorithm. First, based on the characteristics of the infrared image, this study optimizes the network structure of YOLOv7 and proposes a vehicle and pedestrian detection algorithm based on the improved YOLOv7. Aiming at the problem that the reasoning speed and detection accuracy of pedestrian and vehicle detection algorithms based on infrared images are challenging to balance, occupy large storage space, and are difficult to deploy and run in real-time in low and medium-performance devices, the MPConv is added to replace the Conv structure in YOLOv7. In view of the false detection, missed detection, mutual occlusion and overlap of detected objects, and other situations that the YOLOv7 algorithm is prone to cause in the actual deployment environment easily, a tiny object detection layer is added. At the same time, to solve the problem that infrared optical imaging systems are prone to noise caused by external factors, this study introduces the TRPCA method for image denoising in the preprocessing process of the YOLOv7 algorithm. In the end, the HSK-YOLOv7 algorithm is verified using the self-made infrared traffic object detection dataset and the publicly available FLIR dataset to verify the detection effect of the HSK-YOLOv7 algorithm on near-infrared images and thermal infrared images. The parameter quantity of our algorithm is 37.3M, and the computing throughput is 107.5 GFLOPs. The detection speed on the self-made dataset and FLIR dataset reaches 163 frames per second (FPS) and 71.8 FPS, respectively, and the mAP@0.5 indicator reaches 94.08% and 61.3%, respectively. In general, HSK-YOLOv7 can meet the real-time requirements of the autonomous driving system while ensuring detection accuracy.
{"title":"Hold surrounding’s key-you only look once version 7: a real-time pedestrian and vehicle detection algorithm in the low-signal-to-noise ratio infrared image","authors":"Yang Liu, Fulong Yi, Yuhua Ma, Yongfu Wang, Dianhui Wang","doi":"10.1117/1.OE.62.11.118105","DOIUrl":"https://doi.org/10.1117/1.OE.62.11.118105","url":null,"abstract":"Abstract. Compared with visible light imaging technology, infrared optical imaging technology (see Fig. 1) is less affected by weather and illumination and is a potential auxiliary solution method for the future of the autonomous driving system. Based on the above characteristics, this study proposes the hold surrounding’s key (HSK)-you only look once (YOLOv7) algorithm. First, based on the characteristics of the infrared image, this study optimizes the network structure of YOLOv7 and proposes a vehicle and pedestrian detection algorithm based on the improved YOLOv7. Aiming at the problem that the reasoning speed and detection accuracy of pedestrian and vehicle detection algorithms based on infrared images are challenging to balance, occupy large storage space, and are difficult to deploy and run in real-time in low and medium-performance devices, the MPConv is added to replace the Conv structure in YOLOv7. In view of the false detection, missed detection, mutual occlusion and overlap of detected objects, and other situations that the YOLOv7 algorithm is prone to cause in the actual deployment environment easily, a tiny object detection layer is added. At the same time, to solve the problem that infrared optical imaging systems are prone to noise caused by external factors, this study introduces the TRPCA method for image denoising in the preprocessing process of the YOLOv7 algorithm. In the end, the HSK-YOLOv7 algorithm is verified using the self-made infrared traffic object detection dataset and the publicly available FLIR dataset to verify the detection effect of the HSK-YOLOv7 algorithm on near-infrared images and thermal infrared images. The parameter quantity of our algorithm is 37.3M, and the computing throughput is 107.5 GFLOPs. The detection speed on the self-made dataset and FLIR dataset reaches 163 frames per second (FPS) and 71.8 FPS, respectively, and the mAP@0.5 indicator reaches 94.08% and 61.3%, respectively. In general, HSK-YOLOv7 can meet the real-time requirements of the autonomous driving system while ensuring detection accuracy.","PeriodicalId":19561,"journal":{"name":"Optical Engineering","volume":"34 1","pages":"118105 - 118105"},"PeriodicalIF":1.3,"publicationDate":"2023-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139298168","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-31DOI: 10.1117/1.oe.62.10.103104
Jeffrey A. Davis, Ignacio Moreno, Shang Gao, María del Mar Sánchez-López, Don M. Cottrell
{"title":"Multiplexing onto a spatial light modulator using random binary patterns","authors":"Jeffrey A. Davis, Ignacio Moreno, Shang Gao, María del Mar Sánchez-López, Don M. Cottrell","doi":"10.1117/1.oe.62.10.103104","DOIUrl":"https://doi.org/10.1117/1.oe.62.10.103104","url":null,"abstract":"","PeriodicalId":19561,"journal":{"name":"Optical Engineering","volume":"13 70","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135870246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-28DOI: 10.1117/1.oe.63.4.041205
Yousef M. Shishter, Falah H. Ali, Rupert C. Young
Free space diffraction causes the spreading of the received energy at the receiver and thus reduces the signal-to-noise ratio. Bessel–Gauss (BG) beams are considered physically realizable beams, which are robust to free space diffraction over finite propagation distances. Non-diffraction beams have proved useful in many applications, such as optical wireless communications (OWC) and non-linear optics. However, in turbulence BG-beams do suffer from turbulence-induced diffraction. The extended Huygens–Fresnel principle is the main tool of analysis under the effect of strong turbulence. However, the extended Rytov theory (ERT) method provides expressions for the small- and large-scale turbulence-induced signal fluctuations and hence is particularly suitable for statistical channel modeling. In this work, application of the ERT to BG-beams propagating through turbulence is carried out. Closed-form expressions for the induced on-axis small- and large-scale log-irradiance variances are derived. The resultant index of scintillation is analyzed. Then, the error performance of OWC is investigated for BG-beams combined with intensity modulation, M-ary phase shift keying, polarization shift keying, and single-input-multiple-output systems. Significant performance gains are reported compared to Gaussian beams.
{"title":"Scintillation and bit error rate analysis of zero-order Bessel–Gauss beams in atmospheric turbulence based on the extended Rytov theory","authors":"Yousef M. Shishter, Falah H. Ali, Rupert C. Young","doi":"10.1117/1.oe.63.4.041205","DOIUrl":"https://doi.org/10.1117/1.oe.63.4.041205","url":null,"abstract":"Free space diffraction causes the spreading of the received energy at the receiver and thus reduces the signal-to-noise ratio. Bessel–Gauss (BG) beams are considered physically realizable beams, which are robust to free space diffraction over finite propagation distances. Non-diffraction beams have proved useful in many applications, such as optical wireless communications (OWC) and non-linear optics. However, in turbulence BG-beams do suffer from turbulence-induced diffraction. The extended Huygens–Fresnel principle is the main tool of analysis under the effect of strong turbulence. However, the extended Rytov theory (ERT) method provides expressions for the small- and large-scale turbulence-induced signal fluctuations and hence is particularly suitable for statistical channel modeling. In this work, application of the ERT to BG-beams propagating through turbulence is carried out. Closed-form expressions for the induced on-axis small- and large-scale log-irradiance variances are derived. The resultant index of scintillation is analyzed. Then, the error performance of OWC is investigated for BG-beams combined with intensity modulation, M-ary phase shift keying, polarization shift keying, and single-input-multiple-output systems. Significant performance gains are reported compared to Gaussian beams.","PeriodicalId":19561,"journal":{"name":"Optical Engineering","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136161059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-27DOI: 10.1117/1.oe.62.10.106102
Hui Zhou, Wangman Li, Yuan Tan, Zhigao Deng, Ming Chen
{"title":"Cost-efficient radio over fiber system based on 21-tuple frequency binary phase-shift keying millimeter-wave generation without precoding","authors":"Hui Zhou, Wangman Li, Yuan Tan, Zhigao Deng, Ming Chen","doi":"10.1117/1.oe.62.10.106102","DOIUrl":"https://doi.org/10.1117/1.oe.62.10.106102","url":null,"abstract":"","PeriodicalId":19561,"journal":{"name":"Optical Engineering","volume":"274 1-2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136262618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-27DOI: 10.1117/1.oe.63.3.031003
Bin Liu, Hanqing Song, Yan Wang, Lei Liu, Mingguang Shan, Zhi Zhong, Lei Yu
We propose a white-light interferometric demodulation algorithm for high-finesse fiber-optic F-P sensors, in order to improve the demodulation accuracy and the dynamic range encountered in traditional demodulation techniques. The interferometric spectral signal of the high-finesse F-P cavity was converted to the frequency domain and then a more accurate cavity length was estimated based on full phase on higher-order components. A detailed theoretical analysis was operated. A high-finesse F-P temperature sensor based on a silicon diaphragm was fabricated and tested to verify the proposed method. The demodulation accuracy increases with the increase of order, and the anti-noise performance is improved. For the third-order component, the optical path difference sensitivity obtained by this algorithm is 0.231 ± 0.0188 μm / ° C, and the average error rate of cavity length demodulation value is 0.0152%. The proposed algorithm is applicable to demodulate the high-finesse F-P cavities in the light source bandwidth of 1525 to 1575 nm, providing high accuracy and improved anti-noise performance.
{"title":"Spectral demodulation method for high fineness F-P sensors","authors":"Bin Liu, Hanqing Song, Yan Wang, Lei Liu, Mingguang Shan, Zhi Zhong, Lei Yu","doi":"10.1117/1.oe.63.3.031003","DOIUrl":"https://doi.org/10.1117/1.oe.63.3.031003","url":null,"abstract":"We propose a white-light interferometric demodulation algorithm for high-finesse fiber-optic F-P sensors, in order to improve the demodulation accuracy and the dynamic range encountered in traditional demodulation techniques. The interferometric spectral signal of the high-finesse F-P cavity was converted to the frequency domain and then a more accurate cavity length was estimated based on full phase on higher-order components. A detailed theoretical analysis was operated. A high-finesse F-P temperature sensor based on a silicon diaphragm was fabricated and tested to verify the proposed method. The demodulation accuracy increases with the increase of order, and the anti-noise performance is improved. For the third-order component, the optical path difference sensitivity obtained by this algorithm is 0.231 ± 0.0188 μm / ° C, and the average error rate of cavity length demodulation value is 0.0152%. The proposed algorithm is applicable to demodulate the high-finesse F-P cavities in the light source bandwidth of 1525 to 1575 nm, providing high accuracy and improved anti-noise performance.","PeriodicalId":19561,"journal":{"name":"Optical Engineering","volume":"7 8","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136261788","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-26DOI: 10.1117/1.oe.62.10.103102
Randall A. Pietersen, Brian M. Robinson, Melissa S. Beauregard, Herbert H. Einstein
When fielding near-surface hyperspectral imaging systems for computer vision applications, raw data from a sensor are often corrected to reflectance before analysis. This research presents an expedient and flexible methodology for performing spectral reflectance estimation using in situ asphalt cement concrete or Portland cement concrete pavement as a reference material. Then, to evaluate this reflectance estimation method’s utility for computer vision applications, four datasets are generated to train machine learning models for material classification: (1) a raw signal dataset, (2) a normalized dataset, (3) a reflectance dataset corrected with a standard reference material (polytetrafluoroethylene), and (4) a reflectance dataset corrected with a pavement reference material. Various machine learning algorithms are trained on each of the four datasets and all converge to excellent training accuracy (>94 % ). Models trained on the raw or normalized signals, however, did not exceed 70% accuracy when tested against new data captured under different illumination conditions, while models trained using either reflectance dataset saw almost no drop between training and testing accuracy. These results quantify the importance of reflectance correction in machine learning workflows using hyperspectral data, while also confirming practical viability of the proposed reflectance correction method for computer vision applications.
{"title":"Expedient hyperspectral reflectance estimation using in situ pavement reference materials","authors":"Randall A. Pietersen, Brian M. Robinson, Melissa S. Beauregard, Herbert H. Einstein","doi":"10.1117/1.oe.62.10.103102","DOIUrl":"https://doi.org/10.1117/1.oe.62.10.103102","url":null,"abstract":"When fielding near-surface hyperspectral imaging systems for computer vision applications, raw data from a sensor are often corrected to reflectance before analysis. This research presents an expedient and flexible methodology for performing spectral reflectance estimation using in situ asphalt cement concrete or Portland cement concrete pavement as a reference material. Then, to evaluate this reflectance estimation method’s utility for computer vision applications, four datasets are generated to train machine learning models for material classification: (1) a raw signal dataset, (2) a normalized dataset, (3) a reflectance dataset corrected with a standard reference material (polytetrafluoroethylene), and (4) a reflectance dataset corrected with a pavement reference material. Various machine learning algorithms are trained on each of the four datasets and all converge to excellent training accuracy (>94 % ). Models trained on the raw or normalized signals, however, did not exceed 70% accuracy when tested against new data captured under different illumination conditions, while models trained using either reflectance dataset saw almost no drop between training and testing accuracy. These results quantify the importance of reflectance correction in machine learning workflows using hyperspectral data, while also confirming practical viability of the proposed reflectance correction method for computer vision applications.","PeriodicalId":19561,"journal":{"name":"Optical Engineering","volume":"41 8","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136381919","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","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：