Analysis of Visual Interpretation of Satellite Data

Svatonova, H.

doi:10.5194/isprsarchives-xli-b2-675-2016

Cited by 13 publications

(12 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In object detection studies with satellite imagery, the visual interpretation approach that benefits from experts' knowledge for the identification of different objects/targets is still widely used. The accuracy of this approach is dependent on the level of expertise and the approach is time consuming due to the manual process [2].…”

Section: Introductionmentioning

confidence: 99%

Comparative Research on Deep Learning Approaches for Airplane Detection from Very High-Resolution Satellite Images

2020

View full text Add to dashboard Cite

Object detection from satellite images has been a challenging problem for many years. With the development of effective deep learning algorithms and advancement in hardware systems, higher accuracies have been achieved in the detection of various objects from very high-resolution (VHR) satellite images. This article provides a comparative evaluation of the state-of-the-art convolutional neural network (CNN)-based object detection models, which are Faster R-CNN, Single Shot Multi-box Detector (SSD), and You Look Only Once-v3 (YOLO-v3), to cope with the limited number of labeled data and to automatically detect airplanes in VHR satellite images. Data augmentation with rotation, rescaling, and cropping was applied on the test images to artificially increase the number of training data from satellite images. Moreover, a non-maximum suppression algorithm (NMS) was introduced at the end of the SSD and YOLO-v3 flows to get rid of the multiple detection occurrences near each detected object in the overlapping areas. The trained networks were applied to five independent VHR test images that cover airports and their surroundings to evaluate their performance objectively. Accuracy assessment results of the test regions proved that Faster R-CNN architecture provided the highest accuracy according to the F1 scores, average precision (AP) metrics, and visual inspection of the results. The YOLO-v3 ranked as second, with a slightly lower performance but providing a balanced trade-off between accuracy and speed. The SSD provided the lowest detection performance, but it was better in object localization. The results were also evaluated in terms of the object size and detection accuracy manner, which proved that large-and medium-sized airplanes were detected with higher accuracy.

show abstract

Section: Introductionmentioning

confidence: 99%

Comparative Research on Deep Learning Approaches for Airplane Detection from Very High-Resolution Satellite Images

2020

View full text Add to dashboard Cite

show abstract

“…We created four land use maps for the years 2008, 2011, 2013, and 2017/18 based on visual interpretation of high-resolution satellite images and on-site land use mapping in the study area in July 2018 [58,59]. Areas were manually assigned to certain land use classes based on similar structures and band characteristics and the on-site land use mapping.…”

Section: Land Use Mapping and Spatial Statistical Analysismentioning

confidence: 99%

“…The visual interpretation was mainly performed using four high-resolution satellite images provided by the DigitalGlobe Foundation, two from the beginning and two from the end of the dry season (Table 1). These images were compared with the images available in Google Earth Pro from 2007 to 2018 to complement land use information where the high-resolution satellite data are incomplete due to clouds and cloud shadows [59]. Due to different recording angles and slight distortions, a correction with the ArcGIS georeferencing tool was necessary so that the road and path junctions of all images fit.…”

Section: Land Use Mapping and Spatial Statistical Analysismentioning

confidence: 99%

“…We have tried to minimize these potential errors by using similar recording dates, cross checks, matching with Google Earth images, and on-site field mapping. The mapping methods involve many subjective decisions such as the exact boundaries between land uses, which are expressed in the results [58,59].…”

Section: Limits Of Land Use Mapping and Spatial Statistical Analysismentioning

confidence: 99%

See 1 more Smart Citation

Land Use Changes and Their Perception in the Hinterland of Barranquilla, Colombian Caribbean

et al. 2019

View full text Add to dashboard Cite

The coastal strip of the western peri-urban area of Barranquilla in the Atlántico Department (Colombia) is experiencing changes in human-environment interactions through infrastructure, residential, and tourism projects in a vulnerable landscape. In the hilly area, fragments of biodiverse tropical dry forest still exist in various states of conservation and degradation. To understand the interrelated social, economic, and ecological transformations in the area, we analyzed land use change on the local scale including the local community’s perception, because the local community is a key actor for sustainable land use. For the analysis of the interrelated social, economic, and ecological processes, we combined visual interpretation of high-resolution satellite imagery, on-site field land use mapping, and a spatial statistical analysis of the distribution of land use classes with in-depth interviews and a participatory GIS workshop, thus benefitting from the complementary methodological strengths of these approaches. The case study is the rural community of El Morro, which exhibits the typical social, economic, and ecological changes of the coastal strip of the western peri-urban area of Barranquilla. The local community perceives a continuous loss of forest area, but observations from on-site field mapping cannot confirm this linear trend. We observed a gradual replacement of traditional land uses such as smallholder agriculture, charcoal production, and cattle breeding by services for tourism, gated community projects for urban dwellers, and infrastructure projects; these spatial developments have several characteristics of rural gentrification. We conclude that the drivers of environmental degradation have changed and the degradation increased. The development projects of external companies have been rejected by the local community and have induced environmental consciousness among community members. Thus, the local community has become an advocate for sustainable land use in the study area.

show abstract

“…At the same time, the ship target characteristics under different image scenes are also considerably different. For example, the contrast between the foreground and background of a ship in a relatively dark environment is poor [4]; in the environment of dense waves, the detection of the ship is greatly disturbed by the wave wake. For the small target ship with a long wake, it is especially difficult to distinguish the positive sample from the wave wake; In an environment with dense clouds, the texture and clarity of the ship are also greatly disturbed.…”

Section: Introductionmentioning

confidence: 99%

Image Enhancement Driven by Object Characteristics and Dense Feature Reuse Network for Ship Target Detection in Remote Sensing Imagery

Tian

Cao²,

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

As the application scenarios of remote sensing imagery (RSI) become richer, the task of ship detection from an overhead perspective is of great significance. Compared with traditional methods, the use of deep learning ideas has more prospects. However, the Convolutional Neural Network (CNN) has poor resistance to sample differences in detection tasks, and the huge differences in the image environment, background, and quality of RSIs affect the performance for target detection tasks; on the other hand, upsampling or pooling operations result in the loss of detailed information in the features, and the CNN with outstanding results are often accompanied by a high computation and a large amount of memory storage. Considering the characteristics of ship targets in RSIs, this study proposes a detection framework combining an image enhancement module with a dense feature reuse module: (1) drawing on the ideas of the generative adversarial network (GAN), we designed an image enhancement module driven by object characteristics, which improves the quality of the ship target in the images while augmenting the training set; (2) the intensive feature extraction module was designed to integrate low-level location information and high-level semantic information of different resolutions while minimizing the computation, which can improve the efficiency of feature reuse in the network; (3) we introduced the receptive field expansion module to obtain a wider range of deep semantic information and enhance the ability to extract features of targets were at different sizes. Experiments were carried out on two types of ship datasets, optical RSI and Synthetic Aperture Radar (SAR) images. The proposed framework was implemented on classic detection networks such as You Only Look Once (YOLO) and Mask-RCNN. The experimental results verify the effectiveness of the proposed method.

show abstract

Analysis of Visual Interpretation of Satellite Data

Cited by 13 publications

References 15 publications

Comparative Research on Deep Learning Approaches for Airplane Detection from Very High-Resolution Satellite Images

Comparative Research on Deep Learning Approaches for Airplane Detection from Very High-Resolution Satellite Images

Land Use Changes and Their Perception in the Hinterland of Barranquilla, Colombian Caribbean

Image Enhancement Driven by Object Characteristics and Dense Feature Reuse Network for Ship Target Detection in Remote Sensing Imagery

Contact Info

Product

Resources

About