Satellite Image Retrieval Research Articles

Retraction Note: Analytic hierarchy process-based automatic feature weight assignment method for content-based satellite image retrieval system

Retraction Note: Analytic hierarchy process-based automatic feature weight assignment method for content-based satellite image retrieval system

Retraction Note to: Content based satellite image retrieval system using fuzzy clustering

Retraction Note to: Content based satellite image retrieval system using fuzzy clustering

An Integrated Model of Summer and Winter for Chlorophyll-a Retrieval in the Pearl River Estuary Based on Hyperspectral Data

Chlorophyll-a (Chla) is an important parameter for water quality. For remote sensing-based methods for the measurement of Chla, in-situ hyperspectral data is crucial for building retrieval models. In the Pearl River Estuary, we used 61 groups of in-situ hyperspectral data and corresponding Chla concentrations collected in July and December 2020 to build a Chla retrieval model that takes the two different seasons and the turbidity of water into consideration. The following results were obtained. (1) Based on the pre-processing techniques for hyperspectral data, it was shown that the first-derivative of 680 nm is the optimal band for the estimation of Chla in the Pearl River Estuary, with R2 > 0.8 and MAPE of 26.03%. (2) To overcome the spectral resolution problem in satellite image retrieval, based on the simulated reflectance from the Sentinel-2 satellite and the shape of the discrete spectral curve, we constructed a multispectral model using the slope difference index method, which reached a R2 of 0.78 and MAPE of 35.21% and can integrate the summer and winter data. (3) The slope difference method applied to the Sentinel-2 image shows better performance than the red-NIR ratio method. Therefore, the method proposed in this paper is practicable for Chla monitoring of coastal waters based on both in-situ data and images.

Satellite image retrieval of random forest (rf-PNN) based probablistic neural network

Satellite image retrieval of random forest (rf-PNN) based probablistic neural network

Satellite Image Retrieval Based on Adaptive Gaussian Markov Random Field Model with Bayes Back-Propagation Neural Network

Satellite Image Retrieval Based on Adaptive Gaussian Markov Random Field Model with Bayes Back-Propagation Neural Network

An efficient content-based satellite image retrieval system for big data utilizing threshold based checking method

An efficient content-based satellite image retrieval system for big data utilizing threshold based checking method

Distributed Content-Based Image Retrieval of Satellite Images on Hadoop

Background: Owing to increased growth in satellite imagery, the development of an architecture that rapidly and efficiently identifies similar images has become crucial. Hadoop has become a de-facto platform for storing large amounts of data. Apache Spark and MapReduce have also become key frameworks for distributed processing of big data. Objective: This paper proposes a novel distributed content-based image retrieval (DCBIR) architecture that leverages the qualities of these engines, which were not utilized in previous studies. Methods: Features of 40 satellite images with sizes greater than 500 MB were indexed, on a 15-node Hadoop cluster with two different databases viz. Neo4J, a graph database, and HBase, a columnar database. Results: Performance and Scalability of both indexing and query phases, along with precision and recall were observed for both databases. Conclusion: Experimental results show that the proposed system can efficiently perform image retrieval on large remote sensing images.

RETRACTED ARTICLE: Analytic hierarchy process-based automatic feature weight assignment method for content-based satellite image retrieval system

RETRACTED ARTICLE: Analytic hierarchy process-based automatic feature weight assignment method for content-based satellite image retrieval system

RETRACTED ARTICLE: Content based satellite image retrieval system using fuzzy clustering

Nowadays, Satellite image retrieval could be a huge issue to induce information for natural disaster management, military target detection, meteorology, urban designing, harm assessment and change detection, etc. Basis on the image substance, content based image retrieval extracts the images that are relevant to the user given query image from massive image databases. Most of the existing image retrieval methods are still incompetent of providing retrieval outcome with elevated retrieval accuracy and not as much of computational intricacy. This paper proposed Fuzzy multi-characteristic clustering technique to realize this goal that is based on Fuzzy logic and clustering. Fuzzy sets used to represent the vagueness occur in user query, similarity measure and image substance. Clustering is an unsupervised method of classification that provides a small amount of control to clustering and improves the clustering performance drastically. The tentative results reveal that our proposed method can achieve significant precision and recall rates with better computational efficiency.

Machine Learning Paradigm towards Content Based Image Retrieval on High Resolution Satellite Images

In the current era, content based image retrieval based on pattern recognition and classification using machine learning paradigm is an innovative way. In order to retrieve high resolution satellite images Support Vector Machine (SVM) a machine learning paradigm is helpful for learning process and for pattern recognition and classification; ensemble methods give better machine learning results. In this paper, SVM based on random subspace and boosting ensemble learning is proposed for very high resolution satellite image retrieval. The learned SVM ensemble model is used to identify the images that most similar informative for active learning. A bias-weighting system is developed to direct the ensemble model to pay more attention on the positive examples than the negative ones. The UCMerced land use satellite image dataset is used for experimental work. Accuracy and error rate are found to be precise. The tentative effects illustrate that the proposed model derived enhanced retrieval accurateness at the optimum level as well as significantly more effective than existing approaches. The proposed method can diminish the gap dimensionality and conquer the difficulty. The comparisons are evaluated by using precision and recall measurements. Comparative analysis observed that the retrieval time for a particular image have been reduced and the precision is increased. The primary aim of this paper is to represent the significance of ensemble learning with support vector machine in efficient retrieval of image.

Massive-scale visual information retrieval towards city residential environment surveillance

Urban residential environment surveillance plays an important role in modern intelligent city. Satellite images have been applied in various fields, and the analysis and processing of satellite images has become an important means to obtain the information perceived by satellites. This paper focuses on city residential environment surveillance based on massive-scale visual information retrieval. Since the shortcomings of low contrast, blurred boundary, large amount of information and susceptibility to noise, the performance of satellite image segmentation is not satisfactory, which will affect residential environment surveillance. We design an improved rough set fuzzy C-means clustering algorithm combined with ant colony algorithm. More specifically, satellite images are classified based on the gradient of pixels according to the indistinguishable relation of the image combined with rough set theory. Then, the traditional fuzzy set-based fuzzy C-means clustering algorithm is applied to the satellite image segmentation technology. Subsequently, the improved algorithm-quantum ant colony algorithm and rough set fuzzy clustering C-means algorithm are combined to achieve accurate segmentation of satellite images. Afterwards, we propose a satellite image retrieval algorithm, which can assist city residential environment surveillance. Comprehensive experiment show that our proposed method is effective and robust in residential environment surveillance.

Deep learning for multi-modal classification of cloud, shadow and land cover scenes in PlanetScope and Sentinel-2 imagery

With the increasing availability of high-resolution satellite imagery it is important to improve the efficiency and accuracy of satellite image indexing, retrieval and classification. Furthermore, there is a need for utilizing all available satellite imagery in identifying general land cover types and monitoring their changes through time irrespective of their spatial, spectral, temporal and radiometric resolutions. Therefore, in this study, we developed deep learning models able to efficiently and accurately classify cloud, shadow and land cover scenes in different high-resolution (<10 m) satellite imagery. Specifically, we trained deep convolutional neural network (CNN) models to perform multi-label classification of multi-modal, high-resolution satellite imagery at the scene level. Multi-label classification at the scene level (a.k.a. image indexing), as opposed to the pixel level, allows for faster performance, higher accuracy (although at the cost of detail) and higher generalizability. We investigated the generalization ability (i.e. cross-dataset and geographic independence) of individual and ensemble CNN models trained on multi-modal satellite imagery (i.e. PlanetScope and Sentinel-2). The models trained on PlanetScope imagery collected over the Amazon performed well when applied to PlanetScope and Sentinel-2 imagery collected over the Wet Tropics of Australia with an F2 score of 0.72 and 0.69, respectively. Similarly, PlanetScope-based CNN models trained on imagery collected over the Wet Tropics of Australia performed well when applied to Sentinel-2 imagery with an F2 score of 0.76, and the reverse scenario resulted in the same F2 score of 0.76. This suggests that our CNN models have high cross-dataset generalization ability and are suitable for classifying cloud, shadow and land cover classes in satellite imagery with resolutions from 3 m (PlanetScope) to 10 m (Sentinel-2). The performance of our CNN models was also comparable to the state-of-the-art methods (i.e. Sen2Cor and MACCS) developed specifically for classifying cloud and shadow classes in Sentinel-2 imagery. Finally, we show the potential of our CNN models to mask cloud and shadow contaminated areas from PlanetScope- and Sentinel-2-derived NDVI time-series.

An Automatic Accurate High-Resolution Satellite Image Retrieval Method

With the growing number of high-resolution satellite images, the traditional image retrieval method has become a bottleneck in the massive application of high-resolution satellite images because of the low degree of automation. However, there are few studies on the automation of satellite image retrieval. This paper presents an automatic high-resolution satellite image accurate retrieval method based on effective coverage (EC) information, which is used to replace the artificial screening stage in traditional satellite image retrieval tasks. In this method, first, we use a convolutional neural network to extract the EC of each satellite image; then, we use an effective coverage grid set (ECGS) to represent the ECs of all satellite images in the library; finally, the satellite image accurate retrieval algorithm is proposed to complete the process of screening images. The performance evaluation of the method is implemented in three regions: Wuhan, Yanling, and Tangjiashan Lake. The large number of experiments shows that our proposed method can automatically retrieve high-resolution satellite images and significantly improve efficiency.

A Novel Method for Satellite Image Retrieval using Semantic Mining and Hashing

Satellite images play an important role for collection of geographical information. However, the use of such images is limited to a greater extent due to its retrieval complexities. Traditional methods using text have also failed to yield desired and time-saving results. Therefore, Content based image retrieval using high semantic features has been developed to overcome problems related to text retrieval and challenges associated with semantic gap. Keeping in consideration the importance of semantic features for CBIR, the proposed method introduces a novel method using Semantic mining and Hashing for fast retrieval and attainment of accurate results. General Terms Remote sensing, Satellites, Data sets, Content based image retrieval (CBIR) and Algorithms.

An Efficient Approach towards Satellite Image Retrieval using Semantic Mining with Hashing

An Efficient Approach towards Satellite Image Retrieval using Semantic Mining with Hashing

Comparison of approximation methods to Kullback–Leibler divergence between Gaussian mixture models for satellite image retrieval

ABSTRACTAs a probabilistic distance between two probability density functions, Kullback–Leibler divergence is widely used in many applications, such as image retrieval and change detection. Unfortunately, for some models, e.g., Gaussian Mixture Models (GMMs), Kullback–Leibler divergence is not analytically tractable. One has to resort to approximation methods. A number of methods have been proposed to address this issue. In this article, we compare seven methods, namely Monte Carlo method, matched bound approximation, product of Gaussians, variational method, unscented transformation, Gaussian approximation and min-Gaussian approximation, for approximating the Kullback–Leibler divergence between two Gaussian mixture models for satellite image retrieval. Two experiments using two public data sets have been performed. The comparison is carried out in terms of retrieval accuracy and computational time.

Dot Diffusion Block Truncation Coding for Satellite Image Retrieval

In present survey it is noticed that the profound interest in research and study of retrieval of satellite images and Image Retrieval on Content Based is grown hugely .Thus, to build the semantic error which is a huge challenge. Also, it prevents wide hosting of image on content based search engines which is now the necessity of CBIR technique. Mostly image search engine depend on human generated data as input query namely tags or annotations. The images get stored in database according to the annotations or tags assigned to them. Thus, tags or annotations are used as inputs by search engines .Thus, to overcome the limitation of annotations based image retrieval an Improved Block Truncation Coding is used which is based on Texture and color features of an image. In this paper, Improved Block Truncation Coding technique is used to fill the semantic gap. Then the similarity features type of search is carried to find different query regions namely desert, coastal, forest, metro from the satellite images. The retrieval technique can be practically applied and compared using different similarity methods .The technique of Improved Block Truncation Coding is described which is one of the most proficient methods used in retrieval of satellite images. Though, in traditional BTC certain redundancy like false contour, inherent artifacts are observed. In order to deal with this an improved BTC with dotdiffusion is applied to the system. Here, a Dot Diffusion is added for retrieving the most relevant images that best match to the query image whereas Dot Diffusion is used to give best quality image with good clearity and save memory need to store in database.

Pattern Extraction in Segmented Satellite Images By Reducing Semantic Gap Using Relevance Feedback Mechanism

Image Segmentation divides a region into sub regions in such a way that the pixels in a sub region are similar. Satellite image segmentation, an application in image segmentation is to extract the patterns using image retrieval systems. Image Retrieval can be of a wide area in pattern recognition. Content-based can be of best form for retrieval of satellite images. Extraction of high-level features from the images in the form of query or keywords and reduces the semantic gap between the low-level features. To produce better accuracy, Relevance Feed-back based CBIR methods are used for reducing the semantic gap. The overall correlation between the taken satellite image and processed image is identified using Correlation Ratio. This paper focuses on the implementation of K-Means Clustering along with RF mechanism checks the similarity between original and extracted pattern.

System refinement for content based satellite image retrieval

We are witnessing a large increase in satellite generated data especially in the form of images. Hence intelligent processing of the huge amount of data received by dozens of earth observing satellites, with specific satellite image oriented approaches, presents itself as a pressing need. Content based satellite image retrieval (CBSIR) approaches have mainly been driven so far by approaches dealing with traditional images. In this paper we introduce a novel approach that refines image retrieval process using the unique properties to satellite images. Our approach uses a Query by polygon (QBP) paradigm for the content of interest instead of using the more conventional rectangular query by image approach. First, we extract features from the satellite images using multiple tiling sizes. Accordingly the system uses these multilevel features within a multilevel retrieval system that refines the retrieval process. Our multilevel refinement approach has been experimentally validated against the conventional one yielding enhanced precision and recall rates.

Comparative assessment of semantic-sensitive satellite image retrieval: simple and context-sensitive Bayesian networks

In recent years, Bayesian networks using unsupervised extracted image features have been applied in many remote sensing information mining systems to enable semantic-sensitive image retrieval. However, a simple Bayesian network insufficiently accounts for the spatial information, that is, the relations among image regions, for the semantic inference process. This drawback significantly impacts the retrieval performance, especially if the utilised features contain no or little spatial information. Therefore, this article proposes a context-sensitive Bayesian network, which infers semantic concepts of image regions based on the spectral and textural characteristics of the regions themselves as well as their contexts, that is, the adjacent regions. In order to compare the context-sensitive Bayesian network with the simple Bayesian network, comprehensive experiments were conducted based on high-resolution multispectral IKONOS imagery. The results show that the incorporation of the image regions' spatial relations not only significantly improves the accuracy of the semantic concepts inference, but also allows more flexibility in choosing the type of low-level features.