On the design of a color image retrieval method based on combined color descriptors and features

Corn (Zea mays L.) is an important agricultural commodity in Indonesia, serving as the second staple food after rice and playing a crucial role in supporting national food security. However, corn production is frequently threatened by sudden outbreaks of pests and diseases, making accurate early detection essential to maintaining yield stability. This study aims to detect maize leaf diseases using five classification algorithms: Support Vector Machine (SVM), Decision Tree, K-Nearest Neighbors (KNN), Random Forest, and Convolutional Neural Network (CNN). These algorithms were tested using a combination of texture and color features, including Gray Level Co-occurrence Matrix (GLCM), Local Binary Pattern (LBP), Hue-Saturation-Value (HSV), and L*a*b*. The dataset used consists of 2,048 maize leaf images classified into four categories: Blight, Common Rust, Gray Leaf Spot, and Healthy, with 512 images per class. Each class was divided into training and testing sets to train and evaluate the classification models. The results show that CNN achieved the highest accuracy of 93.93% when using a complete combination of color and texture features. Meanwhile, SVM also demonstrated high performance, achieving the same accuracy (93.93%) using only the combination of color features (HSV and Lab*). Random Forest and Decision Tree performed best when using color features alone, with accuracies of 89.81% and 87.14%, respectively. These findings indicate that color features have a dominant influence on classification accuracy, and that combining color and texture features can significantly enhance model performance, particularly in CNN architectures. This study contributes to the development of early disease detection systems in precision agriculture.

Fruit classification under varying pose is still a complicated task due to various properties of numerous types of fruit. In this paper we propose fruit classification method with a novel descriptor as a combination of color and texture feature. Color feature is extracted from segmented fruit image using Color Layout Descriptor, while texture feature is extracted using Edge Histogram Descriptor. Support Vector Machine (SVM) with linear and RBF kernel is used as classifier with 10-fold cross validation. The experimental results demonstrated that our descriptor achieves classification accuracy of over 93.09 % for fruit subcategory and 100 % for fruit category from over 4200 images in varying pose.Keywordsfruit classificationcolortextureSVMColor Layout DescriptorEdge Histogram Descriptor

In this paper, an approach based on combined color and texture features to classify raisins is presented. Least squares support vector machine (LSSVM), linear discriminant analysis, and soft independent modeling of class analogy were used to construct classification models. A total of 480 images were captured from four grades of raisin samples by a Basler 601 fc IEEE1394 digital camera, 200 images were randomly selected to create calibration model (training set), and remaining images were used to verify the model (prediction set). Color features and texture features were obtained from two color spaces: red–green–blue and hue–saturation–intensity using histogram method and gray level co-occurrence matrix method, respectively. Our results indicate that the best performance with about 95% of average correct answer rate is achieved by LSSVM using combined color and texture features from HSI color space. This result is significantly higher than the performance of solely used color or texture features. The combined color and texture features coupled with a LSSVM classifier are a highly accurate way for raisin quality classification.

A fast and efficient image retrieval system based on color and texture features

Low-level image feature extraction is the basis of content based image retrieval (CBIR) systems. In that process, the usage of more than one descriptors has tremendous impact on the increasing of system accuracy. Based on that fact, in this paper we combined color and texture feature in the feature extraction process, namely Color Layout Descriptor (CLD) for color feature extraction and Edge Histogram Descriptor (EHD) for texture feature extraction. We measure the system performance on retrieving top-5, top-10, top-15, and top-20 relevant images. We successfully demonstrated in the experiment, that the combination of color and texture descriptor might be improved the performance of retrieval system, significantly. In our proposed system, the combination of CLD and EHD reaches 72.82% in accuracy, using adaptive weight in Late Fusion Method.

In this paper, we propose a region-based image retrieval system using EHD (Edge Histogram Descriptor) and CLD (Color Layout Descriptor) of MPEG-7 descriptors. The combined descriptor can efficiently describe edge and color features in terms of sub-image regions. That is, the basic unit for the selection of the region-of-interest (ROI) in the image is the sub-image block of the EHD, which corresponds to 16 (i.e., 4x4) non-overlapping image blocks in the image space. This implies that, to have a one-to-one region correspondence between EHD and CLD, we need to take an 8x8 inverse DCT (IDCT) for the CLD. Experimental results show that the proposed retrieval scheme can be used for image retrieval with the ROI based image retrieval for MPEG-7 indexed images.

In the advancement of remote sensing satellite sensors, a large number of high-resolution satellite images are captured every day. To retrieve the required images from a large database has become a challenge. Here, we have used fused color and texture feature for retrieving remote sensing image. Here, we used HSV Histogram, Color moment and color autocorrelogram for color feature extraction. A wavelet transform is used for texture feature extraction. These combined color and texture are used for indexing using k-means clustering. Manhattan distance is also used for similarity matching. UC Merced Land use Land Cover Dataset has been used for the experiment. The k-means clustering with combined color and texture features has shown better retrieval performance than only color features. Indexing has been done using Manhattan distance and k-means clustering. K-means clustering gives better retrieval performance than Manhattan Distance.

Due to the information technology which is rapidly developing, digital content is becoming increasingly difficult to handle. This include images that are kept on digital cameras, CCTV and medical scanners. Areas such as medical and forensic science are using these databases to do critical tasks which include diagnosing of diseases or identification of criminal suspects. However, to manage and search the similar images from these databases are not an easy task. Content Based Image Retrieval (CBIR) is one of the techniques used to manage and search similar images from a database. The performance of CBIR depends on the low level (Texture, Color and Shape) features. In this paper, a new feature vector to represent the image in terms of low level features and to improve the performance of CBIR is proposed. The proposed approach used texture and color feature namely SFTA-CLD. SFTA-CLD is based on Segmentation-based Fractal Texture Analysis (SFTA) and Color Layout Descriptor (CLD). SFTA-CLD is assessed using Coral image gallery and validated by comparing the performance in terms of average precision with previous CBIR techniques.

Rapid development of multimedia technologies made Content Based Image Retrieval (CBIR) an energetic research area for the researchers of multimedia domain. Texture and color features have been the primal descriptors for images in the field of CBIR. This paper proposed a new CBIR system by combining the both color and texture features. Color Layout Descriptor (CLD) from MPEG-7 is used for the color feature extraction while, Mean, variance, skewness, Kurtosis, energy and entropy are used as texture descriptors. Experiments are performed on Coral Database. The results of the proposed method namely CLD-fos are compared with the four well reputed systems (i.e. SIMPLIcity, Histogram based, FIRM, and Variance Segment etc) from the industry. The results of the CLD-fos demonstrated high accuracy rate than the previous systems during the simulations. The proposed CLD-fos achieved significant performance in terms of accuracy.

Content-based video retrieval is a hot topic on video research subject today. To avoid the drawback of traditional retrieval methods which depend on only one feature such as color, texture, shape, movement, this paper put forward a new method which uses both color layout descriptor and homogeneous texture in description MPEG-7. The method extracts color layout descriptor and homogeneous texture descriptor, afterward these two descriptors are normalized. According to the result of retrieving video feature database, it can be concluded that using both features can increase the accuracy and the result suits more to the human vision need.

Most content-based image retrievals (CBIR) use color as image features. However, image retrieval using color features often gives disappointing results because in many cases, images with similar colors do not have similar content. Color methods incorporating spatial information have been proposed to solve this problem, however, these methods often result in very high dimensions of features which drastically slow down the retrieval speed. In this paper, a method combining both color and texture features of image is proposed to improve the retrieval performance. Given a query, images in the database are firstly ranked using color features. Then, the top ranked images are re-ranked according to their texture features. Results show the second process improves retrieval performance significantly.

In this paper we present the first stage of a new approach to improve the precision and recall of the content-based image retrieval task. To do this, we aim to combine three colour features, RGB and HSV histograms, and MPEG- 7 Colour Layout Descriptor. To perform the combination, we propose to use an approximation based on Borda Voting-Schemes. Under that the Borda Voting-Schemes needs at least three votes to perform the combination, we intend to use the K-Nearest Neighbors methods to select the candidate images, given a query image. In the second stage, we’ll implement our approach using at least three image databases.

Color, texture, and shape act as important information for images in human recognition. For content‐based image retrieval, many studies have combined color, texture, and shape features to improve the retrieval performance. However, there have not been many powerful methods for combining all color, texture, and shape features. This study proposes a content‐based image retrieval method that uses the combined local and global features of color, texture, and shape. The color features are extracted from the color autocorrelogram; the texture features are extracted from the magnitude of a complete local binary pattern and the Gabor local correlation revealing local image characteristics; and the shape features are extracted from singular value decomposition that reflects global image characteristics. In this work, an experiment is performed to compare the proposed method with those that use our partial features and some existing techniques. The results show an average precision that is 19.60% higher than those of existing methods and 9.09% higher than those of recent ones. In conclusion, our proposed method is superior over other methods in terms of retrieval performance.

It is vital for a city to monitor water quality in real time since the quality of water has profound effect on residents’ health. Unfortunately, it is impossible for human to monitor water’s chemical composition frequently, which is an impossibly demanding task, let alone in real time. Thus, taking advantage of a highly efficient system can be an excellent solution. In this paper, we created a system called Real-time Intelligent Monitoring System for Water Quality, which enables a city to be responsive to potential outbreak of contamination and to protect city residents. The system is capable of processing and classifying the data extracted from visual images to considerably save more money and labor. In this case, two features Fast Fourier Transform (FFT) and Color Layout Descriptor (CLD) are introduced for Saliency features and color features respectively. FFT performs well in extracting saliency features and is not computationally intensive; CLD is able to represent the color features with high effectiveness and efficiency. Additionally, this system utilizes Support Vector Machine (SVM) based on such features that needs small size of training sets, trains very fast and can classify floating rubbish and any other scenarios of water pollution with satisfying efficiency. Till now, the accuracy has reached 75%, which encourages us. While the detection performance can be further improved, the efficient features & classifiers would serve as powerful methods to automatically monitor water pollution.

Herein, we propose a new Markov random field (MRF) image segmentation model which aims at combining color and texture features. The model has a multi-layer structure: Each feature has its own layer, called feature layer, where an MRF model is defined using only the corresponding feature. A special layer is assigned to the combined MRF model. This layer interacts with each feature layer and provides the segmentation based on the combination of different features. The uniqueness of our algorithm is that it provides both color only and texture only segmentations as well as a segmentation based on combined color and texture features. The number of classes on feature layers is given by the user but it is estimated on the combined layer. © 2002 IEEE.