Content-Based Image Retrieval using Local Features Descriptors and Bag-of-Visual Words

Alkhawlani, Mohammed M.; Elmogy, Mohammed; El-Bakry, Hazem M.

doi:10.14569/ijacsa.2015.060929

Cited by 27 publications

(22 citation statements)

References 19 publications

(20 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Three types of mid-level feature extraction methods describe image semantics, namely the bag-of-visual-words (BoVW), latent Dirichlet allocation (LDA), and machine learning models. However, in practical applications, the performance of BoVW-based methods relies on the extraction of handcrafted local features (Alkhawlani et al 2015). LDA modeling methods rely on K-means clustering to produce a visual dictionary.…”

Section: Introductionmentioning

confidence: 99%

Object-Oriented Method Combined with Deep Convolutional Neural Networks for Land-Use-Type Classification of Remote Sensing Images

Jin

Zhang

et al. 2019

J Indian Soc Remote Sens

View full text Add to dashboard Cite

Land-use information provides a direct representation of the effect of human activities on the environment, and an accurate and efficient land-use classification of remote sensing images is an important element of land-use and land-cover change research. To solve the problems associated with traditional land-use classification methods (e.g., rapid increase in dimensionality of data, inadequate feature extraction, and low running efficiency), a method that combines object-oriented approach with deep convolutional neural network (COCNN) is presented. First, a multi-scale segmentation algorithm is used to segment images to generate image segmentation regions with high homogeneity. Second, a typical rule set of feature objects is constructed on the basis of the object-oriented segmentation results, and the segmentation objects are classified and extracted to form a training sample set. Third, a convolutional neural network (CNN) model structure is modified to improve classification performance, and the training algorithm is optimized to avoid the overfitting phenomenon that occurs during training using small datasets. Ten land-use types are classified by using the remote sensing images covering the area around Fuxian Lake as an example. By comparing the COCNN method with the method based solely on CNN, precision and kappa index were selected to evaluate the classification accuracy of the two methods. For the COCNN method, on the basis of the classification statistics, precision and kappa index coefficients are 96.2% and 0.96, respectively, which are 8.98% and 0.1 higher than those of the method based solely on CNN. Experimental results show that the COCNN method reasonably and efficiently combines object-oriented and deep learning approaches, thereby effectively solving the problem of the inaccurate classification of typical features with better classification accuracy than the simple use of CNN.

show abstract

Section: Introductionmentioning

confidence: 99%

Object-Oriented Method Combined with Deep Convolutional Neural Networks for Land-Use-Type Classification of Remote Sensing Images

Jin

Zhang

et al. 2019

J Indian Soc Remote Sens

View full text Add to dashboard Cite

show abstract

“…Euclidean Distance, Cramer-von Mises Divergence, Manhattan Distance, Cosine Similarity, Chi-Square Dissimilarity, Jeffrey Divergence, Pearson Correlation Coefficient, and Mahalanobis Distance [36], [37]. The results of searching or retrieval from the images collection are sorted and displayed to users [2].…”

Section: Cbirmentioning

confidence: 99%

SVM Relevance Feedback in HSV Quantization for CBIR

Pardede¹

2018

JCP

View full text Add to dashboard Cite

In this research have implemented SVM relevance feedback technique in HSV quantization for CBIR. The proposed technique has two stages. The first stage performs image retrieval process based on results of distance measurement. The distance measurement used is Jeffrey Divergence with threshold 0.15. The second stage is image retrieval process based on SVM RF prediction model. The SVM RF model is formed based on user-provided feedback images. The users' feedback images are labeled as positive and others are negative. The purpose of this study is to determine the best value of the constant C on the linear kernel and the best value of the constant (C, G) on the RBF kernel. The best value of the constants provided generates the best model of SVM RF in the HSV Quantization method so that improve the performance of the CBIR system. Performance measurements are evaluated based on precision, recall, F-measure, and accuracy values. Based on the results of experiments that conduct on Wang dataset obtained that (C, G) = (2 2.725 , 2 2.725 ) is the best value on the RBF kernel. While C = 2 5.275 is the best value on SVM RF using linear kernel. The best of SVM RF technique is SVM RF using RBF kernel of second feedback. The SVM RF using RBF kernel increases the average precision value by 3.02%, the average recall value increasing amount 171.48%, the average F-Measure value increasing amount 80.34%, while the average accuracy value increasing amount 1.99% compared with the baseline. The SVM RF using RBF kernel obtains the best value on both the average recall value and the average F-Measure value compared to the state-of-the-art. and 296-315), 3 Saturations, step (0-0.2, 0.2-0.7 and 0.7-1.0) and 3 Values, step (0-0.2, 0.2-0.7 and 0.7-1.0), and with the same principle done for other schema. So for HSV Quantization (19,4,5) will produce 19*4*5 = 380 the number of distinct colors, whereas HSV Quantization (8,3,3) will produce 8*3*3 = 72 the number of distinct colors, and other HSV Quantization schemes.

show abstract

“…There is interesting work available in the state-of-the-art regarding content based retrieval [1], [2], skin detection [3], [4], and content based filtering [5]. In [5], the authors propose a method combining evidence including video sequences, key shots, and key frames and evaluating performance with three the social networks.…”

Section: Introductionmentioning

confidence: 99%

Media Content Access: Image-based Filtering

Khan¹,

Alkhalifah²

2018

ijacsa

View full text Add to dashboard Cite

As the content on the internet contains sensitive adult material, filtering and blocking this content is essential for the social and ethical values of the many societies and organizations. In this paper, the content filtering is explored from still images' perspectives. Thus, this article investigates and analyses the content based filtering which can help in the flagging of the images as adult nature or safe images. As the proposed approach is based on the Chroma (colour) based skin segmentation and detection for detecting the objectionable content in images; therefore, the approach proceeds in the direction of the classical Machine Learning approaches and uses the two well-known classifiers: 1) The Random Forest; and 2) the Neural Network. Their fusion is also investigated. Skin colour is analyzed in the YCbCr colour space and in the form of blob analysis. With the "Adult vs. Safe" classification, an Accuracy of 0.88 and the low RMSE of 0.313 is achieved, indicating the usefulness of the detection model.

show abstract

Content-Based Image Retrieval using Local Features Descriptors and Bag-of-Visual Words

Cited by 27 publications

References 19 publications

Object-Oriented Method Combined with Deep Convolutional Neural Networks for Land-Use-Type Classification of Remote Sensing Images

Object-Oriented Method Combined with Deep Convolutional Neural Networks for Land-Use-Type Classification of Remote Sensing Images

SVM Relevance Feedback in HSV Quantization for CBIR

Media Content Access: Image-based Filtering

Contact Info

Product

Resources

About