Pub Date : 2023-08-01DOI: 10.14358/pers.23-00003r2
Yanxiong Liu, Zhipeng Dong, Yikai Feng, Yilan Chen, Long Yang
Edge detection in high-spatial-resolution remote sensing images (HSRIs ) is a key technology for automatic extraction, analysis, and understanding of image information. With respect to the problem of fake edges in image edge detection caused by image noise and the phenomenon of the same class objects reflecting different spectra, this article proposes a novel edge detection method for HSRIs by combin- ing superpixels with dual-threshold edge tracking. First, the image is smoothed using the simple linear iterative clustering algorithm to eliminate the influence of image noise and the phenomenon of the same class objects reflecting different spectra on image edge detec - tion. Second, initial edge detection results of the image are obtained using the dual-threshold edge tracking algorithm. Finally, the initial image edge detection results are post-processed by removing the burrs and extracting skeleton lines to obtain accurate edge detection results. The experimental results confirm that the proposed method outperforms the others and can obtain smooth, continuous, and single-pixel response edge detection results for HSRIs .
{"title":"Edge Detection Method for High-Resolution Remote Sensing Imagery by Combining Superpixels with Dual-Threshold Edge Tracking","authors":"Yanxiong Liu, Zhipeng Dong, Yikai Feng, Yilan Chen, Long Yang","doi":"10.14358/pers.23-00003r2","DOIUrl":"https://doi.org/10.14358/pers.23-00003r2","url":null,"abstract":"Edge detection in high-spatial-resolution remote sensing images (HSRIs ) is a key technology for automatic extraction, analysis, and understanding of image information. With respect to the problem of fake edges in image edge detection caused by image noise and the phenomenon of the\u0000 same class objects reflecting different spectra, this article proposes a novel edge detection method for HSRIs by combin- ing superpixels with dual-threshold edge tracking. First, the image is smoothed using the simple linear iterative clustering algorithm to eliminate the influence of image\u0000 noise and the phenomenon of the same class objects reflecting different spectra on image edge detec - tion. Second, initial edge detection results of the image are obtained using the dual-threshold edge tracking algorithm. Finally, the initial image edge detection results are post-processed\u0000 by removing the burrs and extracting skeleton lines to obtain accurate edge detection results. The experimental results confirm that the proposed method outperforms the others and can obtain smooth, continuous, and single-pixel response edge detection results for HSRIs .","PeriodicalId":211256,"journal":{"name":"Photogrammetric Engineering & Remote Sensing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128147827","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}
{"title":"GIS Automated Delineation of Hospital Service Areas by Fahui Wang and Changzhen Wang","authors":"D. Zourarakis","doi":"10.14358/pers.89.8.465","DOIUrl":"https://doi.org/10.14358/pers.89.8.465","url":null,"abstract":"","PeriodicalId":211256,"journal":{"name":"Photogrammetric Engineering & Remote Sensing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128061186","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 : 2023-08-01DOI: 10.14358/pers.23-00001r2
Z. Shao, Muhammad Nasar Ahmad, Akib Javed, F. Islam, Zahid Jahangir, Israr Ahmad
Impervious surfaces are an essential component of our environment and are mainly triggered by human developments. Rapid urbanization and population expansion have increased Lahore's urban impervious surface area. This research is based on estimating the urban imper- vious surface area ( uisa ) growth from 1993 to 2022. Therefore, we aimed to generate an accurate urban impervious surfaces area map based on Landsat time series data on Google Earth Engine ( gee ). We have used a novel global impervious surface area index ( gisai ) for impervious surface area ( uisa ) extraction. The gisai accomplished significant results, with an average overall accuracy of 90.93% and an average kappa coefficient of 0.78. We also compared the results of gisai with Global Human Settlement Layer-Built and harmonized nighttime light ( ntl ) isa data products. The accuracy assessment and cross-validation of uisa results were performed using ground truth data on ArcGIS and gee. Our research findings revealed that the spatial extent of uisa increased by 198.69 km2 from 1993 to 2022 in Lahore. Additionally, the uisa has increased at an average growth rate of 39.74 km2. The gisai index was highly accurate at extract- ing uisa and can be used for other cities to map impervious surface area growth. This research can help urban planners and policymak- ers to delineate urban development boundaries. Also, there should be controlled urban expansion policies for sustainable metropolis and should use less impermeable materials for future city developments.
不透水的表面是我们环境的重要组成部分，主要是由人类发展引发的。快速的城市化和人口扩张增加了拉合尔的城市不透水地表面积。本研究基于对1993年至2022年城市不透水表面积(uisa)增长的估计。因此，我们的目标是基于谷歌地球引擎(gee)上的Landsat时间序列数据生成精确的城市不透水地表地图。我们使用一种新的全球不透水面指数(gisai)来提取不透水面(uisa)。gisai取得了显著的结果，平均总体准确率为90.93%，平均kappa系数为0.78。我们还将gisai的结果与Global Human Settlement Layer-Built and harmonized night light (ntl)数据产品进行了比较。使用ArcGIS和gee上的地面真值数据对usisa结果进行准确性评估和交叉验证。研究结果表明，1993 ~ 2022年拉合尔城市土地利用面积增加了198.69 km2。此外，美国的平均增长率为39.74平方公里。gisai指数在提取地形图时具有很高的准确性，可用于其他城市绘制不透水地表面积的增长图。该研究可为城市规划者和决策者划定城市发展边界提供参考。此外，应该有控制的城市扩张政策，以实现可持续的大都市，应该在未来的城市发展中使用更少的不透水材料。
{"title":"Expansion of Urban Impervious Surfaces in Lahore (1993–2022) Based on Gee and Remote Sensing Data","authors":"Z. Shao, Muhammad Nasar Ahmad, Akib Javed, F. Islam, Zahid Jahangir, Israr Ahmad","doi":"10.14358/pers.23-00001r2","DOIUrl":"https://doi.org/10.14358/pers.23-00001r2","url":null,"abstract":"Impervious surfaces are an essential component of our environment and are mainly triggered by human developments. Rapid urbanization and population expansion have increased Lahore's urban impervious surface area. This research is based on estimating the urban imper- vious surface area\u0000 ( uisa ) growth from 1993 to 2022. Therefore, we aimed to generate an accurate urban impervious surfaces area map based on Landsat time series data on Google Earth Engine ( gee ). We have used a novel global impervious surface area index ( gisai ) for impervious surface area ( uisa ) extraction.\u0000 The gisai accomplished significant results, with an average overall accuracy of 90.93% and an average kappa coefficient of 0.78. We also compared the results of gisai with Global Human Settlement Layer-Built and harmonized nighttime light ( ntl ) isa data products. The accuracy assessment\u0000 and cross-validation of uisa results were performed using ground truth data on ArcGIS and gee. Our research findings revealed that the spatial extent of uisa increased by 198.69 km2 from 1993 to 2022 in Lahore. Additionally, the uisa has increased at an average growth rate of 39.74\u0000 km2. The gisai index was highly accurate at extract- ing uisa and can be used for other cities to map impervious surface area growth. This research can help urban planners and policymak- ers to delineate urban development boundaries. Also, there should be controlled urban expansion\u0000 policies for sustainable metropolis and should use less impermeable materials for future city developments.","PeriodicalId":211256,"journal":{"name":"Photogrammetric Engineering & Remote Sensing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122660441","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 : 2023-08-01DOI: 10.14358/pers.23-00004r3
Jiaxin Xu, Qiao Zhang, Yu Liu, Mengting Zheng
For remorte sensing image object detection tasks in the small object feature, extraction ability is insufficient and difficult to locate, and other problems. This paper proposes an improved algorithm for small object detection in remote sensing images based on a window self-attention mechanism. On the basis of You Only Look Once (YOLO)v5s, a shallow feature extraction layer with four times downsampling is added to the feature fusion pyramid and the window self-attention mechanism is added to the Path Aggregation Network. Experiments show that the improved model obtained the Mean Average Precision (mAP) of 78.3% and 91.8% on the DIOR and Remote Sensing Object Detection public data sets with frames per second of 65 and 51, respectively. Compared with the basal YOLOv5s network, the mAP has improved by 5.8% and 3.3%, respectively. Compared with other object detection methods, the detection accuracy and real-time performance have been improved.
对于遥感图像目标检测任务中存在的小目标特征、提取能力不足、难以定位等问题。提出了一种基于窗口自关注机制的遥感图像小目标检测改进算法。在YOLO (You Only Look Once)v5s的基础上，在特征融合金字塔中增加了四次下采样的浅层特征提取层，在路径聚合网络中增加了窗口自关注机制。实验表明，在每秒帧数为65帧的DIOR和每秒帧数为51帧的遥感目标检测公开数据集上，改进模型的Mean Average Precision (mAP)分别达到78.3%和91.8%。与基础YOLOv5s网络相比，mAP分别提高了5.8%和3.3%。与其他目标检测方法相比，提高了检测精度和实时性。
{"title":"Small Object Detection in Remote Sensing Images Based on Window Self-Attention Mechanism","authors":"Jiaxin Xu, Qiao Zhang, Yu Liu, Mengting Zheng","doi":"10.14358/pers.23-00004r3","DOIUrl":"https://doi.org/10.14358/pers.23-00004r3","url":null,"abstract":"For remorte sensing image object detection tasks in the small object feature, extraction ability is insufficient and difficult to locate, and other problems. This paper proposes an improved algorithm for small object detection in remote sensing images based on a window self-attention\u0000 mechanism. On the basis of You Only Look Once (YOLO)v5s, a shallow feature extraction layer with four times downsampling is added to the feature fusion pyramid and the window self-attention mechanism is added to the Path Aggregation Network. Experiments show that the improved model obtained\u0000 the Mean Average Precision (mAP) of 78.3% and 91.8% on the DIOR and Remote Sensing Object Detection public data sets with frames per second of 65 and 51, respectively. Compared with the basal YOLOv5s network, the mAP has improved by 5.8% and 3.3%, respectively. Compared with other object detection\u0000 methods, the detection accuracy and real-time performance have been improved.","PeriodicalId":211256,"journal":{"name":"Photogrammetric Engineering & Remote Sensing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129896549","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}
{"title":"SectorInsight.com —The Journey from a Ph.D. to a Successful Company","authors":"K. Lim","doi":"10.14358/pers.89.7.401","DOIUrl":"https://doi.org/10.14358/pers.89.7.401","url":null,"abstract":"","PeriodicalId":211256,"journal":{"name":"Photogrammetric Engineering & Remote Sensing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115442653","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 : 2023-07-01DOI: 10.14358/pers.22-00106r2
E. Gonsoroski, Y. Ahn, E. Harville, Nathaniel Countess, M. Lichtveld, K. Pan, L. Beitsch, S. Sherchan, C. Uejio
Post-hurricane damage assessments are often costly and time-consuming. Remotely sensed data provides a complementary method of data collection that can be completed comparatively quickly and at relatively low cost. This study focuses on 15 Florida counties impacted by Hurricane Michael (2018), which had category 5 strength winds at landfall. The present study evaluates the ability of aerial imagery collected to cost-effectively measure blue tarps on buildings for disaster impact and recovery. A support vector machine model classified blue tarp, and parcels received a damage indicator based on the model's prediction. The model had an overall accuracy of 85.3% with a sensitivity of 74% and a specificity of 96.7%. The model results indicated approximately 7% of all parcels (27 926 residential and 4431 commercial parcels) in the study area as having blue tarp present. The study results may benefit jurisdictions that lacked financial resources to conduct on-the-ground damage assessments.
{"title":"Classifying Building Roof Damage Using High Resolution Imagery for Disaster Recovery","authors":"E. Gonsoroski, Y. Ahn, E. Harville, Nathaniel Countess, M. Lichtveld, K. Pan, L. Beitsch, S. Sherchan, C. Uejio","doi":"10.14358/pers.22-00106r2","DOIUrl":"https://doi.org/10.14358/pers.22-00106r2","url":null,"abstract":"Post-hurricane damage assessments are often costly and time-consuming. Remotely sensed data provides a complementary method of data collection that can be completed comparatively quickly and at relatively low cost. This study focuses on 15 Florida counties impacted by Hurricane Michael\u0000 (2018), which had category 5 strength winds at landfall. The present study evaluates the ability of aerial imagery collected to cost-effectively measure blue tarps on buildings for disaster impact and recovery. A support vector machine model classified blue tarp, and parcels received a damage\u0000 indicator based on the model's prediction. The model had an overall accuracy of 85.3% with a sensitivity of 74% and a specificity of 96.7%. The model results indicated approximately 7% of all parcels (27 926 residential and 4431 commercial parcels) in the study area as having blue tarp present.\u0000 The study results may benefit jurisdictions that lacked financial resources to conduct on-the-ground damage assessments.","PeriodicalId":211256,"journal":{"name":"Photogrammetric Engineering & Remote Sensing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114807598","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}
{"title":"GIS Tips & Tricks — File Navigation is Easier than You Think","authors":"Alma M. Karlin","doi":"10.14358/pers.89.7.407","DOIUrl":"https://doi.org/10.14358/pers.89.7.407","url":null,"abstract":"","PeriodicalId":211256,"journal":{"name":"Photogrammetric Engineering & Remote Sensing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124590612","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 : 2023-07-01DOI: 10.14358/pers.23-00006r2
R. Sultanova, R. Mustafin
At the research points, the relationship between the Normalized Difference Vegetation and Normalized Green Red Difference indices is characterized by a determination coefficient equal to 0.52. The estimation of the emission of carbon dioxide and nitrogen oxide in the forest air at an altitude of 40 m above the level of the soil cover during the growing season showed differences in their values during the daytime and at night. The results helped determine promising methods of inventory of the carbon landfill forest area for aboveground woody biomass assessment based on data obtained from several sources and land forest estimation research. The research involved: 1) integration of an unmanned aerial vehicle -based digital camera and lidar sensors in order to optimize the efficiency and cost of data collection; 2) taking advantage of high-resolution aerial photographs and sparse lidar point clouds using an information fusion approach and the ability to compensate for their shortcomings.
{"title":"Estimation of the Forest Stand Biomass and Greenhouse Gas Emissions Using Lidar Surveys","authors":"R. Sultanova, R. Mustafin","doi":"10.14358/pers.23-00006r2","DOIUrl":"https://doi.org/10.14358/pers.23-00006r2","url":null,"abstract":"At the research points, the relationship between the Normalized Difference Vegetation and Normalized Green Red Difference indices is characterized by a determination coefficient equal to 0.52. The estimation of the emission of carbon dioxide and nitrogen oxide in the forest air at an\u0000 altitude of 40 m above the level of the soil cover during the growing season showed differences in their values during the daytime and at night. The results helped determine promising methods of inventory of the carbon landfill forest area for aboveground woody biomass assessment based on\u0000 data obtained from several sources and land forest estimation research. The research involved: 1) integration of an unmanned aerial vehicle -based digital camera and lidar sensors in order to optimize the efficiency and cost of data collection; 2) taking advantage of high-resolution aerial\u0000 photographs and sparse lidar point clouds using an information fusion approach and the ability to compensate for their shortcomings.","PeriodicalId":211256,"journal":{"name":"Photogrammetric Engineering & Remote Sensing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115282180","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 : 2023-07-01DOI: 10.14358/pers.22-00130r2
Linfeng Wu, Huajun Wang, Huiqing Wang
Deep learning (dl), especially convolutional neural networks (cnns), has been proven to be an excellent feature extractor and widely applied to hyperspectral image (hsi) classification. However, dl is a computationally demanding algorithm with many parameters and a high computational burden, which seriously restricts the deployment of dl-based hsi classification algorithms on mobile and embedded systems. In this paper, we propose an extremely lightweight conditional three-dimensional (3D) hsi with a double-branch structure to solve these problems. Specifically, we introduce a lightweight conditional 3D convolution to replace the conventional 3D convolution to reduce the computational and memory cost of the network and achieve flexible hsi feature extraction. Then, based on lightweight conditional 3D convolution, we build two parallel paths to independently exploit and optimize the diverse spatial and spectral features. Furthermore, to precisely locate the key information, which is conducive to classification, a lightweight attention mechanism is carefully designed to refine extracted spatial and spectral features, and improve the classification accuracy with less computation and memory costs. Experiments on three public hsi data sets show that the proposed model can effectively reduce the cost of computation and memory, achieve high execution speed, and better classification performance compared with several recent dl-based models.
{"title":"A Lightweight Conditional Convolutional Neural Network for Hyperspectral Image Classification","authors":"Linfeng Wu, Huajun Wang, Huiqing Wang","doi":"10.14358/pers.22-00130r2","DOIUrl":"https://doi.org/10.14358/pers.22-00130r2","url":null,"abstract":"Deep learning (dl), especially convolutional neural networks (cnns), has been proven to be an excellent feature extractor and widely applied to hyperspectral image (hsi) classification. However, dl is a computationally demanding algorithm with many parameters and a high computational\u0000 burden, which seriously restricts the deployment of dl-based hsi classification algorithms on mobile and embedded systems. In this paper, we propose an extremely lightweight conditional three-dimensional (3D) hsi with a double-branch structure to solve these problems. Specifically,\u0000 we introduce a lightweight conditional 3D convolution to replace the conventional 3D convolution to reduce the computational and memory cost of the network and achieve flexible hsi feature extraction. Then, based on lightweight conditional 3D convolution, we build two parallel paths\u0000 to independently exploit and optimize the diverse spatial and spectral features. Furthermore, to precisely locate the key information, which is conducive to classification, a lightweight attention mechanism is carefully designed to refine extracted spatial and spectral features, and improve\u0000 the classification accuracy with less computation and memory costs. Experiments on three public hsi data sets show that the proposed model can effectively reduce the cost of computation and memory, achieve high execution speed, and better classification performance compared with several\u0000 recent dl-based models.","PeriodicalId":211256,"journal":{"name":"Photogrammetric Engineering & Remote Sensing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123478031","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}
{"title":"Book Review: Fundamentals of Capturing and Processing Drone Imagery and Data. Edited by Amy E. Frazier and Kunwar K. Singh","authors":"Kathryn M. Rocheford","doi":"10.14358/pers.89.7.405","DOIUrl":"https://doi.org/10.14358/pers.89.7.405","url":null,"abstract":"","PeriodicalId":211256,"journal":{"name":"Photogrammetric Engineering & Remote Sensing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129434754","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：