Color and Texture Feature Based Scene Classification

2020 ◽  
Vol 17 (11) ◽  
pp. 4897-4901
Author(s):  
V. Gayathri ◽  
Eric Clapten ◽  
S. Mahalakshmi ◽  
S. Rajes Kannan
Keyword(s):  
Land Use ◽  
Feature Extraction ◽  
Texture Feature ◽  
Rotation Invariance ◽  
Scene Classification ◽  
Work Of Art ◽  
Image Brightness ◽  
Spatial Features ◽  
Neighborhood Function ◽  
Feature Based

Right now, overall trademark based multiscale multiresolution multistructure (M3LBP) neighborhood parallel example and nearby characteristic based totally min blend feature extraction is proposed for scene category. To extract international functions, characterize the leading spatial features in a couple of scale, a couple of choice, more than one structure way. The micro/macro shape facts and rotation invariance are guaranteed inside the worldwide function extraction approach. Neighborhood function extraction, coloration histogram characteristic (CHF) can thoroughly explain the spatial coloration statistics of an image. It also describes the image brightness, color statistics of a photo, which encompass the picture coloration distribution, photo assessment. The CHF can be computed from the min max shade quantizes. Ultimately Fused feature instance amongst nearby and international capabilities because the scene descriptor to prepare a portion based absolutely extreme finding a workable pace for scene style is outfitted. The proposed strategy is radically assessed on benchmark scene datasets (the 21 magnificence land use scene), and the trial results show that the proposed procedure prompts predominant kind standard execution as contrasted and the realm of-work of art style systems.

scholarly journals Spectral–Spatial Feature Partitioned Extraction Based on CNN for Multispectral Image Compression

2020 ◽  
Vol 13 (1) ◽  
pp. 9
Author(s):  
Fanqiang Kong ◽  
Kedi Hu ◽  
Yunsong Li ◽  
Dan Li ◽  
Shunmin Zhao
Keyword(s):  
Feature Extraction ◽  
Image Compression ◽  
Multispectral Imaging ◽  
Rapid Development ◽  
Spectral Feature ◽  
Multispectral Image ◽  
Landsat 8 ◽  
Spatial Feature ◽  
Spatial Features ◽  
Multispectral Image Compression

Recently, the rapid development of multispectral imaging technology has received great attention from many fields, which inevitably involves the image transmission and storage problem. To solve this issue, a novel end-to-end multispectral image compression method based on spectral–spatial feature partitioned extraction is proposed. The whole multispectral image compression framework is based on a convolutional neural network (CNN), whose innovation lies in the feature extraction module that is divided into two parallel parts, one is for spectral and the other is for spatial. Firstly, the spectral feature extraction module is used to extract spectral features independently, and the spatial feature extraction module is operated to obtain the separated spatial features. After feature extraction, the spectral and spatial features are fused element-by-element, followed by downsampling, which can reduce the size of the feature maps. Then, the data are converted to bit-stream through quantization and lossless entropy encoding. To make the data more compact, a rate-distortion optimizer is added to the network. The decoder is a relatively inverse process of the encoder. For comparison, the proposed method is tested along with JPEG2000, 3D-SPIHT and ResConv, another CNN-based algorithm on datasets from Landsat-8 and WorldView-3 satellites. The result shows the proposed algorithm outperforms other methods at the same bit rate.

scholarly journals A Multi-Branch Feature Fusion Strategy Based on an Attention Mechanism for Remote Sensing Image Scene Classification

2021 ◽  
Vol 13 (10) ◽  
pp. 1950
Author(s):  
Cuiping Shi ◽  
Xin Zhao ◽  
Liguo Wang
Keyword(s):  
Remote Sensing ◽  
Feature Extraction ◽  
Classification Accuracy ◽  
Feature Fusion ◽  
State Of The Art ◽  
Rapid Development ◽  
Remote Sensing Image ◽  
Classification Performance ◽  
Attention Mechanism ◽  
Scene Classification

In recent years, with the rapid development of computer vision, increasing attention has been paid to remote sensing image scene classification. To improve the classification performance, many studies have increased the depth of convolutional neural networks (CNNs) and expanded the width of the network to extract more deep features, thereby increasing the complexity of the model. To solve this problem, in this paper, we propose a lightweight convolutional neural network based on attention-oriented multi-branch feature fusion (AMB-CNN) for remote sensing image scene classification. Firstly, we propose two convolution combination modules for feature extraction, through which the deep features of images can be fully extracted with multi convolution cooperation. Then, the weights of the feature are calculated, and the extracted deep features are sent to the attention mechanism for further feature extraction. Next, all of the extracted features are fused by multiple branches. Finally, depth separable convolution and asymmetric convolution are implemented to greatly reduce the number of parameters. The experimental results show that, compared with some state-of-the-art methods, the proposed method still has a great advantage in classification accuracy with very few parameters.

Spatial features of land use/land cover change in the United States

2003 ◽  
Vol 13 (1) ◽  
pp. 63-70 ◽  
Author(s):  
Zhiqiang Gao ◽  
Jiyuan Liu ◽  
Xiangzheng Deng
Keyword(s):  
United States ◽  
Land Use ◽  
Land Cover ◽  
Land Cover Change ◽  
The United States ◽  
Land Use Land Cover ◽  
Spatial Features

Contour- and Texture-based analysis for victim identification in forensic odontology

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
G. Jaffino ◽  
J. Prabin Jose
Keyword(s):  
Feature Extraction ◽  
Texture Feature ◽  
Forensic Odontology ◽  
Individual Identification ◽  
Texture Feature Extraction ◽  
Person Identification ◽  
Content Type ◽  
Photographic Images ◽  
Shape Extraction ◽  
Contour Shape

PurposeForensic dentistry is the application of dentistry in legal proceedings that arise from any facts relating to teeth. The ultimate goal of forensic odontology is to identify the individual when there are no other means of identification such as fingerprint, Deoxyribonucleic acid (DNA), iris, hand print and leg print. The purpose of selecting dental record is for the teeth to be able to withstand decomposition, heat degradation up to 1600 °C. Dental patterns are unique for every individual. This work aims to analyze the contour shape extraction and texture feature extraction of both radiographic and photographic dental images for person identification.Design/methodology/approachTo achieve an accurate identification of individuals, the missing tooth in the radiograph has to be identified before matching of ante-mortem (AM) and post-mortem (PM) radiographs. To identify whether the missing tooth is a molar or premolar, each tooth in the given radiograph has to be classified using a k-nearest neighbor (k-NN) classifier; then, it is matched with the universal tooth numbering system. In order to make exact person identification, this research work is mainly concentrate on contour shape extraction and texture feature extraction for person identification. This work aims to analyze the contour shape extraction and texture feature extraction of both radiographic and photographic images for individual identification. Then, shape matching of AM and PM images is performed by similarity and distance metric for accurate person identification.FindingsThe experimental results are analyzed for shape and feature extraction of both radiographic and photographic dental images. From this analysis, it is proved that the higher hit rate performance is observed for the active contour shape extraction model, and it is well suited for forensic odontologists to identify a person in mass disaster situations.Research limitations/implicationsForensic odontology is a branch of human identification that uses dental evidence to identify the victims. In mass disaster circumstances, contours and dental patterns are very useful to extract the shape in individual identification.Originality/valueThe experimental results are analyzed both the contour shape extraction and texture feature extraction of both radiographic and photographic images. From this analysis, it is proved that the higher hit rate performance is observed for the active contour shape extraction model and it is well suited for forensic odontologists to identify a person in mass disaster situations. The findings provide theoretical and practical implications for individual identification of both radiographic and photographic images with a view to accurate identification of the person.

Sign in / Sign up

Export Citation Format

Share Document