APPLICATION OF SEMANTIC SEGMENTATION WITH FEW LABELS IN THE DETECTION OF WATER BODIES FROM PERUSAT-1 SATELLITE’S IMAGES

Abstract. Remote sensing is widely used to monitor earth surfaces with the main objective of extracting information from it. Such is the case of water surface, which is one of the most affected extensions when flood events occur, and its monitoring helps in the analysis of detecting such affected areas, considering that adequately defining water surfaces is one of the biggest problems that Peruvian authorities are concerned with. In this regard, semiautomatic mapping methods improve this monitoring, but this process remains a time-consuming task and into the subjectivity of the experts.In this work, we present a new approach for segmenting water surfaces from satellite images based on the application of convolutional neural networks. First, we explore the application of a U-Net model and then a transfer knowledge-based model. Our results show that both approaches are comparable when trained using an 680-labelled satellite image dataset; however, as the number of training samples is reduced, the performance of the transfer knowledge-based model, which combines high and very high image resolution characteristics, is improved.

Download Full-text

A Comparative Analysis of Existing Satellite Image Enhancement Techniques for Effective Visual Display

Journal of Computational and Theoretical Nanoscience ◽

10.1166/jctn.2019.8285 ◽

2019 ◽

Vol 16 (9) ◽

pp. 4003-4007 ◽

Cited By ~ 2

Author(s):

Neetu Manocha ◽

Rajeev Gupta

Keyword(s):

Comparative Analysis ◽

Image Enhancement ◽

High Frequency ◽

Satellite Images ◽

Satellite Image ◽

Research Work ◽

Visual Display ◽

Image Resolution ◽

High Image Resolution

Due to environment untidiness and inappropriate setting or dealing of camera, a satellite image contains blur or other types of noises. These images are captured by satellites consist lots of information about the surface of earth or other planets. But, due to blur or noise, the quality of these images is degraded. Now days, there are many fields in which satellite images are used, which effects the environment. The accuracy and effective visual display of satellite images with high image resolution using CBIR technique is major concern. This paper presents a comparative analysis of existing satellite image enhancement techniques to reduce the blur of an image on the basis of accuracy and response time. The aim of research work is to eliminate the noise without losing high frequency details and to enhance the image for effective visual display.

Download Full-text

Towards Sub-Pixel Automatic Geometric Corrections of Very-High Resolution Panchromatic Satellite Data of Urban Areas

Remote Sensing ◽

10.3390/rs11091097 ◽

2019 ◽

Vol 11 (9) ◽

pp. 1097 ◽

Cited By ~ 1

Author(s):

Aleš Marsetič ◽

Peter Pehani

Keyword(s):

Feature Extraction ◽

High Resolution ◽

Urban Areas ◽

Satellite Images ◽

Reference Data ◽

Geometric Model ◽

Control Point ◽

Satellite Image ◽

Extraction Algorithm ◽

Very High

This paper presents an automatic procedure for the geometric corrections of very-high resolution (VHR) optical panchromatic satellite images. The procedure is composed of three steps: an automatic ground control point (GCP) extraction algorithm that matches the linear features that were extracted from the satellite image and reference data; a geometric model that applies a rational function model; and, the orthorectification procedure. Accurate geometric corrections can only be achieved if GCPs are employed to precisely correct the geometric biases of images. Due to the high resolution and the varied acquisition geometry of images, we propose a fast, segmentation based method for feature extraction. The research focuses on densely populated urban areas, which are very challenging in terms of feature extraction and matching. The proposed algorithm is capable of achieving results with a root mean square error of approximately one pixel or better, on a test set of 14 panchromatic Pléiades images. The procedure is robust and it performs well in urban areas, even for images with high off-nadir angles.

Download Full-text

THE RECONSTRUCTION OF SATELLITE IMAGES BASED ON FRACTAL INTERPOLATION

Fractals ◽

10.1142/s0218348x11005385 ◽

2011 ◽

Vol 19 (03) ◽

pp. 347-354 ◽

Cited By ~ 2

Author(s):

CHING-JU CHEN ◽

SHU-CHEN CHENG ◽

Y. M. HUANG

Keyword(s):

High Resolution ◽

Satellite Images ◽

Satellite Image ◽

Image Data ◽

Image Resolution ◽

Source Image ◽

Fractal Interpolation ◽

Low Resolution ◽

Resolution Image ◽

High Resolution Image

This study discussed the application of a fractal interpolation method in satellite image data reconstruction. It used low-resolution images as the source data for fractal interpolation reconstruction. Using this approach, a high-resolution image can be reconstructed when there is only a low-resolution source image available. The results showed that the high-resolution image data from fractal interpolation can effectively enhance the sharpness of the border contours. Implementing fractal interpolation on an insufficient image resolution image can avoid jagged edges and mosaic when enlarging the image, as well as improve the visibility of object features in the region of interest. The proposed approach can thus be a useful tool in land classification by satellite images.

Download Full-text

A new approach to the application of conflict redistribution rule in Satellite Image Classification

Ukrainian journal of remote sensing ◽

10.36023/ujrs.2020.25.171 ◽

2020 ◽

pp. 12-16

Author(s):

Sofiia Alpert

Keyword(s):

Remote Sensing ◽

Image Classification ◽

Satellite Images ◽

Satellite Image ◽

Classification Methods ◽

New Approach ◽

Remote Sensing Classification ◽

Satellite Image Classification ◽

Bodies Of Evidence

Nowadays solution of different scientific problems using satellite images, generally includes a classification procedure. Classification is one of the most important procedures used in remote sensing, because it involves a lot of mathematical operations and data preprocessing. The processing of information and combining of conflicting data is a very difficult problem in classification tasks. Nowadays many classification methods are applied in remote sensing. Classification of conflicting data has been a key problem, both from a theoretical and practical point of view. But a lot of known classification methods can not deal with highly conflicted data and uncertainty. The main purpose of this article is to apply proportional conflict redistribution rule (PRC5) for satellite image classification in conditions of uncertainty, when conflicting sources of evidence give incomplete and vague information. This rule can process conflicting data and combine conflicting bodies of evidence (spectral bands). Proportional conflict redistribution rule can redistribute the partial conflicting mass proportionally on non-empty sets involved in the conflict. It was noticed, that this rule can provide a construction of aggregated estimate under conflict. It calculates all partial conflicting masses separately. It was also shown, that proportional conflict redistribution rule is the most mathematically exact redistribution of conflicting mass to non-empty set. But this rule consists of difficult calculation procedures. The more hypotheses and more masses are involved in the fusion, the more difficult is to implement proportional conflict redistribution rule, therefore special computer software should be used. It was considered an example of practical use of the proposed conflict redistribution rule. It also was noticed, that this new approach to the application of conflict redistribution rule in satellite image classification can be applied for analysis of satellite images, solving practical and ecological tasks, assessment of agricultural lands, classification of forests, in searching for oil and gas.

Download Full-text

Mapping and Area Estimation of Mango Orchards of Lucknow Region by Applying Knowledge Based Decision Tree to Landsat 8 OLI Satellite Images

International Journal of Innovative Technology and Exploring Engineering - Special Issue ◽

10.35940/ijitee.b8109.019320 ◽

2020 ◽

Vol 9 (3) ◽

pp. 3627-3635

Keyword(s):

Decision Tree ◽

Satellite Images ◽

Vegetation Index ◽

Accuracy Assessment ◽

Satellite Image ◽

Accurate Information ◽

Landsat 8 ◽

Atmospheric Conditions ◽

Area Estimation ◽

Knowledge Based

Mango is a very important fruit which is liked by majority of the population due to its nutritional value and excellent taste. India is the largest producer of mango in the world. Accurate information is required for policy decision making in terms of providing subsidy, area expansion, and crop insurance planning. Hence, this type of information may be retrieve through satellite images by using the image classification techniques, which are playing a crucial role in crop cover classification, yield prediction and crop monitoring etc. Classification of optical satellite images is still a challenging task due to effect of changing atmospheric conditions such as cloud, snow, haze, dust, fog, and rain etc. In this paper, knowledge based decision tree classification (DTC) has been proposed to classify the mango orchards of Lucknow district using multi-temporal Landsat 8 operational land imager (OLI) images from year 2015 to 2017 and further mango orchard area were also estimated. In order to develop the DTC, separability analysis for various land cover classes was carried out on different vegetation indices namely, normalized difference vegetation index (NDVI), modified normalized difference water index (MNDWI), and soil adjusted vegetation index (SAVI). In order to analyze the performance of DTC, most commonly used satellite image classifiers such as unsupervised classifier (i.e. ISODATA) and supervised classifier (i.e. Maximum Likelihood) have been used and it is observed that the proposed DTC outperformed these traditional classifiers. Also, accuracy assessment has been carried out to measure the performance of proposed DTC and it is observed that all of the three images from 2015 to 2017 are classified with high overall accuracy, which is ranging from 70.66% to 86.69%. Kappa Coefficient (KC) for all the three images ranged from 0.65 to 0.83, which indicates that classified images are highly acceptable for area estimation.

Download Full-text

CLOUD AND CLOUD SHADOW MASKING USING MULTI-TEMPORAL CLOUD MASKING ALGORITHM IN TROPICAL ENVIRONMENTAL

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xli-b2-95-2016 ◽

2016 ◽

Vol XLI-B2 ◽

pp. 95-100 ◽

Cited By ~ 3

Author(s):

D. S. Candra ◽

S. Phinn ◽

P. Scarth

Keyword(s):

Satellite Images ◽

Confusion Matrix ◽

Satellite Image ◽

Landsat 8 ◽

Qualitative And Quantitative ◽

Algorithm Comparison ◽

Multi Temporal ◽

The Difference ◽

Very High ◽

Better Than

A cloud masking approach based on multi-temporal satellite images is proposed. The basic idea of this approach is to detect cloud and cloud shadow by using the difference reflectance values between clear pixels and cloud and cloud shadow contaminated pixels. Several bands of satellite image which have big difference values are selected for developing Multi-temporal Cloud Masking (MCM) algorithm. Some experimental analyses are conducted by using Landsat-8 images. Band 3 and band 4 are selected because they can distinguish between cloud and non cloud. Afterwards, band 5 and band 6 are used to distinguish between cloud shadow and clear. The results show that the MCM algorithm can detect cloud and cloud shadow appropriately. Moreover, qualitative and quantitative assessments are conducted using visual inspections and confusion matrix, respectively, to evaluate the reliability of this algorithm. Comparison between this algorithm and QA band are conducted to prove the reliability of the approach. The results show that MCM better than QA band and the accuracy of the results are very high.

Download Full-text

MSLp: Deep Superresolution for Meteorological Satellite Image

Complexity ◽

10.1155/2021/2678124 ◽

2021 ◽

Vol 2021 ◽

pp. 1-8

Author(s):

Liling Zhao ◽

Hao Yu ◽

Yan Wang

Keyword(s):

High Resolution ◽

Satellite Images ◽

Weather Forecasting ◽

Satellite Image ◽

Image Resolution ◽

Complex Data ◽

Opinion Score ◽

Phase Loss ◽

High Resolution Images ◽

Meteorological Satellite

High-resolution meteorological satellite image is the basic data for weather forecasting, climate prediction, and early warning of various meteorological disasters. However, the poor image resolution is limited for both subjective and automated analyses. Through our investigation and study, we found that the meteorological satellite image is a kind of complex data with multimodal and multitemporal characteristics. Fortunately, based on zero-shot learning theory, the complexity of the meteorological satellite image can be used to enhance its own image resolution. In this work, we propose a novel framework called MSLp (Meteorological Satellite Loss phase). Specifically, we choose a zero-shot network as a backbone and propose a phase loss function. A mapping from low- to high-resolution meteorological satellite images was trained for improving the resolution by up to a factor of 4×. Our quantitative study demonstrates the superiority of the proposed approach over ZSSR and bicubic interpolation. For qualitative analysis, visual tests were performed by 7 meteorologists to confirm the utility of the proposed algorithm. The mean opinion score is 9.32 (the full score is 10). These meteorologists think that weather forecasters need higher-resolution meteorological satellite images and the high-resolution images obtained by our method have the potential to be a great help for weather analysis and forecasting.

Download Full-text

Discussion of different Remote sensing satellite possibilities for scientifical Earth observations

E3S Web of Conferences ◽

10.1051/e3sconf/202126404007 ◽

2021 ◽

Vol 264 ◽

pp. 04007

Author(s):

Aybek Arifjanov ◽

Shamshodbek Akmalov ◽

Shakhzod Shodiev ◽

Abdukarim Haitov

Keyword(s):

High Resolution ◽

Satellite Images ◽

Water Discharge ◽

Satellite Image ◽

Research Area ◽

Daily Monitoring ◽

Medium Resolution ◽

Very High Spatial Resolution ◽

High Resolution Images ◽

Very High

More than 1,000 satellites are launched into space, and they differ in their functions, rotation orbits, resolution, and other properties. Scientists divide the satellites into low-resolution, medium-resolution, high-resolution, and very high-resolution satellites by their properties. Now, the biggest challenge facing scientists is to use some of these different resolution images in their field. To get the expected result, it is very important to analyze the image that needs an which gives more accurate results. Therefore, the main attention of this article is aimed to find the answer to these problems. In this article 3 satellite images which have different resolution are analyzed. The possibility of middle-resolution images of MODIS, high-resolution images of Landsat, and very high-resolution images of WorldView-2 (WV-2) satellites using GIS are analyzed. A research area was the Syrdarya region, and downloaded different images of satellites of this area and compared with using e Cognition. According to the results, a more accurate satellite image for irrigation sets information is WorldView-2 images. In comparison analysis, it shows more accurate properties than other satellite images. As irrigation sets are small objects for the analysis, very high spatial resolution satellite images are important. Water discharge and surface change happen very fast; thus, it requires daily monitoring of the condition. And in this case, the temporal resolution of the MODIS and Landsat is 16 day, and it is a too long period.

Download Full-text

Tree-based convolutional neural networks for object classification in segmented satellite images

The International Journal of High Performance Computing Applications ◽

10.1177/1094342020945026 ◽

2020 ◽

pp. 109434202094502 ◽

Cited By ~ 2

Author(s):

Y Harold Robinson ◽

S Vimal ◽

Manju Khari ◽

Fernando Carlos López Hernández ◽

Rubén González Crespo

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Satellite Images ◽

Semantic Representation ◽

Computing Time ◽

Computational Cost ◽

Semantic Segmentation ◽

Extraction Process ◽

Set Up ◽

Very High

Satellite images have a very high resolution, which make their automatic processing computationally costly, and they suffer from artifacts making their processing difficult. This paper describes a method for the effective semantic segmentation of satellite images, and compares different object classifiers in terms of accuracy and execution time. In the paper, the image spectrum is used to reduce the computational cost during the segmentation and classification steps. Firstly, artifacts are corrected from the satellite images for facilitating the feature extraction process. After this, semantic representation is used to gather the semantic regions of downscaled images. As the images are very large, this scaling down significantly reduces the computing time with little degradation in the coarse object detection results. A deep neural forest classifier finds potential regions before executing the pixel-based segmentation. Finally, in our experiments, boundary detection and several classifiers are evaluated to find the objects associated with these regions. The paper details the set-up for our tree-based convolutional neural network. The results indicate that this tree-based convolutional neural network outperforms the other surveyed techniques in the literature.

Download Full-text