Article
Geochemistry & Geophysics
Armin Moghimi, Ali Mohammadzadeh, Turgay Celik, Meisam Amani
Summary: This article proposes a new relative radiometric normalization method for multitemporal satellite images, utilizing optimization and image fusion strategies. Experimental results show that the method outperforms the state-of-the-art methods, reducing the average root-mean-square error by up to 32% across all data sets.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Hanzeyu Xu, Yuyu Zhou, Yuchun Wei, Houcai Guo, Xiao Li
Summary: Relative radiometric normalization (RRN) is an effective method for enhancing radiometric consistency among multitemporal satellite images. In this study, we propose a multirule-based RRN method that identifies spectral- and spatial-invariant pseudo-invariant features (PIFs) and uses partial least-squares (PLS) regression to model RRN, resulting in improved radiometric consistency between reference-target image pairs. Our method outperforms six other RRN methods and shows potential for generating more comparable bitemporal multisensor images.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Engineering, Electrical & Electronic
Hanzeyu Xu, Yuyu Zhou, Yuchun Wei, Chong Liu, Xiao Li, Wei Chen
Summary: In this study, a novel RRN method was proposed to enhance the radiometric consistency of Landsat time-series images by trend-based PIFs identification, PIFs optimization, and combined RRN modeling. The experimental results showed that the proposed method achieved good performance in model precision and radiance consistency improvement, outperforming seven commonly used RRN methods.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Wessel Bonnet, Turgay Celik
Summary: A novel RRN method based on random sample consensus is proposed in this letter, which can effectively perform radiometric calibration and resist the influence of anomalous pixels. Experimental results demonstrate that this method outperforms existing RRN methods in all metrics considered in this study.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Armin Moghimi, Amin Sarmadian, Ali Mohammadzadeh, Turgay Celik, Meisam Amani, Huseyin Kusetogullari
Summary: In this study, a novel method for radiometric correction of unregistered multisensor image pairs is proposed. The method utilizes the KAZE detector to extract feature points and applies the conditional probability process in linear model fitting for correction. Experimental results demonstrate that the proposed method outperforms existing methods both qualitatively and quantitatively, indicating its high efficiency.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Armin Moghimi, Turgay Celik, Ali Mohammadzadeh, Huseyin Kusetogullari
Summary: This article compares the performance of commonly used keypoint detectors and descriptors in keypoint-based relative radiometric normalization of unregistered bitemporal multispectral images. The keypoint-based RRN is shown to be robust against variations in spatial resolution, illumination, and sensors. Blob detectors are more accurate but computationally expensive compared to corner detectors in RRN.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Environmental Sciences
Armin Moghimi, Ali Mohammadzadeh, Turgay Celik, Brian Brisco, Meisam Amani
Summary: Relative radiometric normalization (RRN) is crucial for pre-processing and analyzing multitemporal remote sensing (RS) images. This study proposes a new automatic RRN technique that selects clustered pseudo-invariant features (PIFs) and uses fusion-based modeling to improve the accuracy and efficiency of RRN results.
Article
Geochemistry & Geophysics
Jie Han, Zui Tao, Yong Xie, Huina Li, Xiaoguo Guan, Hang Yi, Tingting Shi, Gengke Wang
Summary: This study addresses the insufficiency of cross-calibration methods in the wide field of view imaging system of the GaoFen-6 satellite. By using MODIS as a reference sensor, cross-calibration methods based on the TOA and BOA BRDF models are developed. The results show that the BOA BRDF model provides higher consistency with the official calibration coefficients in the cross-calibration of eight spectral bands.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Rui Liu, Feng Wang, Niangang Jiao, Wei Yu, Hongjian You, Fangjian Liu
Summary: This study proposes a radiometric normalization method based on the radiometric principle, which considers the radiometric characteristic of SAR images and optimizes all images as a whole to reduce the radiometric differences between SAR mosaic images.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Engineering, Civil
Aya Selmoune, Jeongin Yun, Myoungkook Seo, Hyeokhyeon Kwon, Changhee Lee, Jinwoo Lee
Summary: Pedestrians are at a higher risk of serious injury in vehicle collisions, especially on residential roads without dividers and with blind spots. Traditional safety features are not always effective in preventing accidents. To address this issue, a collision risk warning service using CCTVs and radar to detect objects in real-time has been proposed. User surveys conducted on university campus roads showed that the service was considered necessary and significantly contributed to traffic safety.
JOURNAL OF ADVANCED TRANSPORTATION
(2023)
Article
Geochemistry & Geophysics
Yongkun Liu, Tengfei Long, Weili Jiao, Guojin He, Bo Chen, Peng Huang
Summary: This article proposes a general relative radiometric correction method based on the gray-level co-occurrence matrix (GLCM), which can effectively correct vignetting, chromatic aberration, and other problems in Level-0 images captured by spaceborne push-broom imaging system. The experiments show that the proposed method has good robustness and universality.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Remote Sensing
Armin Moghimi, Turgay Celik, Ali Mohammadzadeh
Summary: This paper proposes a robust algorithm for radiometric normalization of bitemporal multispectral images, addressing the limitations of existing methods. The proposed algorithm utilizes a new extension of SURF detector for multispectral images and a flexible switching regression model. Experimental results demonstrate that the proposed method outperforms conventional methods in terms of accuracy and visual quality.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2022)
Article
Chemistry, Analytical
Gabriel Yedaya Immanuel Ryadi, Muhammad Aldila Syariz, Chao-Hung Lin
Summary: Multitemporal cross-sensor imagery is important for monitoring the Earth's surface, but it often lacks visual consistency due to atmospheric and surface variations. A relaxation-based algorithm is proposed to normalize satellite images, which improves radiometric consistency and maintains important features.
Article
Economics
Zagros Z. Dilshad, Swar O. Ahmed, Bootan Rahman
Summary: This study examines the effects of the COVID-19 pandemic on employment in the Kurdistan Region Government, including lower wages and job loss. The findings show that the crisis has led to salary cuts, business shutdowns, and reduced working hours. There is no association between losing salary and working status, but there is a correlation between gender and business shutdown.
JOURNAL OF THE KNOWLEDGE ECONOMY
(2023)
Article
Computer Science, Artificial Intelligence
Jingyang Wang, Jiazheng Li, Xiaoxiao Wang, Jue Wang, Min Huang
Summary: A new method of CT-LSTM is proposed in this study to establish an air quality prediction model by combining chi-square test and long short-term memory network model. The experimental results demonstrate that this new method has high accuracy and low error, outperforming the other four methods in predicting AQI levels.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Environmental Sciences
Xi Li, Li Yan, Lijun Lu, Guoman Huang, Zheng Zhao, Zechang Lu
Summary: This study used SBAS and NSBAS techniques to monitor land subsidence in the Hebei Plain, China, and presented a novel data fusion flow for generating subsidence velocity. Results identified 26 typical subsidence bowls, with the highest subsidence velocity observed in Gaoyang County, Baoding City. The typical subsidence bowls showed high spatial correlation with shallow and deep groundwater funnels, different land use types, and faults.
Article
Environmental Sciences
Li Yan, Jianbing Yang, Yi Zhang
Summary: Detection of ground surface changes is crucial in many fields. We propose a new architecture that achieves building change detection and instance segmentation simultaneously, achieving state-of-the-art performance on two datasets.
JOURNAL OF THE INDIAN SOCIETY OF REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Jian Wang, Li Yan, Keming Yang, Wei Tang, Hong Xie, Shuyi Yao, Zhihua Xu, Jianbing Yang
Summary: This study proposes a novel method based on InSAR to accurately estimate mining-induced 3-D surface deformations. The method integrates single InSAR interferogram, the Gompertz time function, and the probability integral model to derive the deformations at any moment. Experimental results demonstrate the accuracy of the method in subsidence, tilt, curvature, displacement, and strain, and its ability to accurately estimate deformations under different geological conditions.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Xi Li, Li Yan, Yi Zhang, Nan Mo
Summary: This letter proposes a deep-supervised dual discriminative metric network (SDMNet) for high-resolution remote sensing image change detection. It utilizes a discriminative decoder network and a discriminative implicit metric module to aggregate contextual information and measure the distance between features, resulting in accurate and robust CD results. The method also introduces multiple change graph losses for deep supervision.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)