Color Feature Extraction Research Articles

Research on lane detection algorithms based on GTNs

Abstract. Lane detection technology plays an important role in the lane departure warning and adaptive cruise control functions of autonomous vehicle systems. Traditional lane line detection includes experimental steps such as preyreatment, color space conversion, feature extraction, and lane line tracking, but there are still some problems such as recognition accuracy in complex environments and parameter adjustment dependency. To overcome these limitations, this study uses a novel method that can directly predict the parameters of the lane shape model, thus avoiding complex post-processing steps. This research method uses the network structure based on Graph Transformer Networks (GTNS) to capture richer structural information and contextual relationships, thereby improving the accuracy of detection. In terms of feature extraction, Graph Transformer Networks is used to accurately capture the structural information of the lane through the attention mechanism. At the same time, the results in the later stage show that our method has better advantages in accuracy and speed. In short, this optimization scheme not only improves the detection speed but also ensures a certain degree of accuracy. The method will perform better in the future after further optimization such as multi-sensor fusion and real-time optimization.

Revisiting representation learning of color information: Color medical image segmentation incorporating quaternion

Currently, color medical image segmentation methods commonly extract color and texture features mixed together by default, however, the distribution of color information and texture information is different: color information is represented differently in different color channels of a color image, while the distribution of texture information remains the same. Such a simple and brute-force feature extraction pattern will inevitably result in a partial bias in the model's semantics understanding. In this paper, we decouple the representation learning for color-texture information, and propose a novel network for color medical image segmentation, named CTNet. Specifically, CTNet introduces the Quaternion CNN (QCNN) module to capture the correlation among different color channels of color medical images to generate color features, and uses a designed local-global texture feature integrator (LoG) to mine the textural features from local to global. Moreover, a multi-stage features interaction strategy is proposed to minimize the semantic understanding gap of color and texture features in CTNet, so that they can be subsequently fused to generate a unified and robust feature representation. Comparative experiments on four different color medical image segmentation benchmark datasets show that CTNet strikes an optimal trade-off between segmentation accuracy and computational overhead when compared to current state-of-the-art methods. We also conduct extensive ablation experiments to verify the effectiveness of the proposed components. Our code will be available at https://github.com/Notmezhan/CTNet.

DDNet: Depth Dominant Network for Semantic Segmentation of RGB-D Images.

Convolutional neural networks (CNNs) have been widely applied to parse indoor scenes and segment objects represented by color images. Nonetheless, the lack of geometric and context information is a problem for most RGB-based methods, with which depth features are only used as an auxiliary module in RGB-D semantic segmentation. In this study, a novel depth dominant network (DDNet) is proposed to fully utilize the rich context information in the depth map. The critical insight is that obvious geometric information from the depth image is more conducive to segmentation than RGB data. Compared with other methods, DDNet is a depth-based network with two branches of CNNs to extract color and depth features. As the core of the encoder network, the depth branch is given a larger fusion weight to extract geometric information, while semantic information and complementary geometric information are provided by the color branch for the depth feature maps. The effectiveness of our proposed depth-based architecture has been demonstrated by comprehensive experimental evaluations and ablation studies on challenging RGB-D semantic segmentation benchmarks, including NYUv2 and a subset of ScanNetv2.

Klasifikasi Tingkat Kemanisan Buah Kersen Berdasarkan Fitur Warna NTSC Menggunakan Jaringan Syaraf Tiruan Berbasis Pengolahan Citra Digital

<em><span lang="EN-US">The fruit of the calabura tree (Muntingia calabura) is a small red fruit originating from the Prunus genus, often found along roadsides. This fruit contains numerous nutrients beneficial for bodily health, serving as a highly potential source of nutrition. Presently, a challenge exists in determining the sweetness level of calabura fruit, relying heavily on manual human assessment. The development of classification utilizing technology is considered a crucial step. Previous research has concentrated on classifying various objects using RGB, HSV, YCbCr color feature extraction. However, it was observed that RGB, HSV, YCbCr color features are not universally suitable, particularly for calabura fruits. Hence, this study employs a method of classifying the sweetness level of calabura fruit based on NTSC color features using a Digital Image Processing-based Artificial Neural Network (ANN). This approach leverages color-based image processing features. The research involves several stages, starting from acquiring 300 calabura fruit images with 3 levels of classification to the classification process utilizing Backpropagation in the ANN. Multiple training and testing scenarios were conducted to select feature combinations with the highest accuracy and fastest computational time. Results revealed that the most effective feature used was the NTSC color feature as a skin characteristic parameter. Based on training outcomes using 210 training images, the accuracy reached 100% with a computational time of 1.66 seconds per image. Meanwhile, testing with 90 sample images showed an accuracy of 94% with a computational time of 4.23 seconds per image. Thus, it can be concluded that the employed method successfully classifies the quality of calabura fruit images based on color features and skin characteristics.</span></em>

K-means 알고리즘을 기반으로 한 야오족 자수 색상 추출 및 가상 캐릭터 의상디자인 응용

To promote the digital preservation and transmission of traditional cultural elements, this study focuses on Yao embroidery as a case study, combining digital technologies and algorithms to extract and analyze its color characteristics. These extracted color features were applied to the design of virtual character clothing. First, a bilateral filtering algorithm was used to preprocess the images of Yao embroidery, reducing noise while preserving intricate details. Subsequently, Euclidean distance was calculated in the Lab color space, and the OSTU adaptive binarization method was employed to separate the image from its background. Additionally, K-means clustering was applied multiple times to the color data, and the optimal result was achieved when the number of clusters was set to eight. Through this process, eight major colors of Yao embroidery were extracted. Based on these extracted colors, 3D point cloud analysis was conducted to study the three-dimensional spatial distribution of color points, and a color combination model based on hue, brightness, and saturation was constructed. The results revealed that the color composition of Yao embroidery predominantly consists of high-saturation colors, with strong color contrasts and distinct ethnic characteristics. Ultimately, this study demonstrated the applicability of integrating the color model and design patterns of Yao embroidery into virtual character clothing design, showcasing the potential for incorporating traditional embroidery into digital design. This contributes to the preservation and promotion of intangible cultural heritage and serves as a valuable reference for further research on the color systems of traditional ethnic costumes.

Identifikasi Kematangan Buah Pisang Berdasarkan Variasi Jarak Menggunakan Metode K-Nearest Neighbor

This research aims to identify the level of ripeness of kepok bananas based on the color of their skin using the K-Nearest Neighbor (K-NN) method. Bananas are an important commodity in Indonesia, and various ripeness levels need to be identified. The current process of identifying banana ripeness is still done manually, which requires a lot of labor and tends to be subjective. The K-NN method is used to classify bananas based on their skin color. This research involves the collection of banana images with three ripeness levels (raw, ripe, and overripe) and the extraction of RGB color features from these images. Three distance methods, namely Euclidean, Minkowski, and Manhattan, are also employed to compare accuracy results. The evaluation results of this research show that the accuracy value for the Euclidean distance method is 84%, the Minkowski distance method is 82%, and the Manhattan distance method is 80%. Thus, the findings indicate that the K-NN method and the Euclidean distance method provide good results in identifying the ripeness level of bananas. By implementing the K-NN algorithm, this research attempts to address the weaknesses of the time-consuming and subjective manual identification process, with the hope of providing a more accurate and efficient solution for the banana industry. The results of this research can be used to automate the identification process of banana ripeness levels and improve efficiency in banana sorting. It is expected that this research can provide practical benefits to the community and serve as a basis for further research in this field.

Multi-Feature Extraction and Selection Method to Diagnose Burn Depth from Burn Images

Burn wound depth is a significant determinant of patient treatment. Typically, the evaluation of burn depth relies heavily on the clinical experience of doctors. Even experienced surgeons may not achieve high accuracy and speed in diagnosing burn depth. Thus, intelligent burn depth classification is useful and valuable. Here, an intelligent classification method for burn depth based on machine learning techniques is proposed. In particular, this method involves extracting color, texture, and depth features from images, and sequentially cascading these features. Then, an iterative selection method based on random forest feature importance measure is applied. The selected features are input into the random forest classifier to evaluate this proposed method using the standard burn dataset. This method classifies burn images, achieving an accuracy of 91.76% when classified into two categories and 80.74% when classified into three categories. The comprehensive experimental results indicate that this proposed method is capable of learning effective features from limited data samples and identifying burn depth effectively.

CCM-Net: Color compensation and coordinate attention guided underwater image enhancement with multi-scale feature aggregation

Due to the light scattering and wavelength absorption in water, underwater images exhibit blurred details, low contrast, and color deviation. Existing underwater image enhancement methods are divided into traditional methods and deep learning-based methods. Traditional methods either rely on scene prior and lack robustness, or are not flexible enough resulting in poor enhancement effects. Deep learning methods have achieved good results in the field of underwater image enhancement due to their powerful feature representation ability. However, these methods cannot enhance underwater images with various degradations because they do not consider the inconsistent attenuation of different color channels and spatial regions. In this paper, we propose a novel asymmetric encoder-decoder network for underwater image enhancement, called CCM-Net. Concretely, we first introduce the prior knowledge-based encoder, which includes color compensation (CC) modules and feature extraction modules that consist of depth-wise separable convolution and global-local coordinate attention (GLCA). Then, we design a multi-scale feature aggregation (MFA) module to integrate shallow, middle, and deep features. Finally, we deploy a decoder to reconstruct the underwater images with the extracted features. Extensive experiments on publicly available datasets demonstrate that our CCM-Net effectively improves the visual quality of underwater images and achieves impressive performance.

A lightweight color and geometry feature extraction and fusion module for end-to-end 6D pose estimation

Although advancements in red–green–blue-depth (RGB-D)-based six degree-of-freedom (6D) pose estimation methods, severe occlusion remains challenging. Addressing this issue, we propose a novel feature fusion module that can efficiently leverage the color and geometry information in RGB-D images. Unlike prior fusion methods, our method employs a two-stage fusion process. Initially, we extract color features from RGB images and integrate them into a point cloud. Subsequently, an anisotropic separable set abstraction network-like network is utilized to process the fused point cloud, extracting both local and global features, which are then combined to generate the final fusion features. Furthermore, we introduce a lightweight color feature extraction network to reduce model complexity. Extensive experiments conducted on the LineMOD, Occlusion LineMOD, and YCB-Video datasets conclusively demonstrate that our method significantly enhances prediction accuracy, reduces training time, and exhibits robustness to occlusion. Further experiments show that our model is significantly smaller than the latest popular 6D pose estimation models, which indicates that our model is easier to deploy on mobile platforms.

Predictive modeling of rice milling degree for three typical Chinese rice varieties using interpretative machine learning methods.

Brown rice over-milling causes high economic and nutrient loss. The rice degree of milling (DOM) detection and prediction remain a challenge for moderate processing. In this study, a self-established grain image acquisition platform was built. Degree of bran layer remaining (DOR) datasets is established with image capturing and processing (grain color, texture, and shape features extraction). The mapping relationship between DOR and the DOM is in-depth analyzed. Rice grain DOR typical machine learning and deep learning prediction models are established. The results indicate that the optimized Catboost model can be established with cross-validation and grid search method, with the best accuracy improving from 84.28% to 91.24%, achieving precision 91.31%, recall 90.89%, and F1-score 91.07%. Shapley additive explanations analysis indicates that color, texture, and shape feature affect Catboost prediction accuracy, the feature importance: color>texture>shape. The YCbCr-Cb_ske and GLCM-Contrast features make the most significant contribution to rice milling quality prediction. The feature importance provides theoretical and practical guidancefor grain DOM prediction model. PRACTICAL APPLICATION: Rice milling degree prediction and detection are valuable for rice milling process in practical application. In this paper, image processing and machine learning methods provide an automated, nondestructive, and cost-effective way to predict the quality of rice. The study may serve as a valuable reference for improving rice milling methods, retaining rice nutrition, and reducing broken rice yield.

Asymmetric lesion detection with geometric patterns and CNN-SVM classification

In dermoscopic images, which allow visualization of surface skin structures not visible to the naked eye, lesion shape offers vital insights into skin diseases. In clinically practiced methods, asymmetric lesion shape is one of the criteria for diagnosing Melanoma. Initially, we labeled data for a non-annotated dataset with symmetrical information based on clinical assessments. Subsequently, we propose a supporting technique—a supervised learning image processing algorithm—to analyze the geometrical pattern of lesion shape, aiding non-experts in understanding the criteria of an asymmetric lesion. We then utilize a pre-trained convolutional neural network (CNN) to extract shape, color, and texture features from dermoscopic images for training a multiclass support vector machine (SVM) classifier, outperforming state-of-the-art methods from the literature. In the geometry-based experiment, we achieved a 99.00 % detection rate for dermatological asymmetric lesions. In the CNN-based experiment, the best performance is found 94 % Kappa Score, 95 % Macro F1-score, and 97 % weighted F1-score for classifying lesion shapes (Asymmetric, Half-Symmetric, and Symmetric).

Application of Attention-Enhanced 1D-CNN Algorithm in Hyperspectral Image and Spectral Fusion Detection of Moisture Content in Orah Mandarin (Citrus reticulata Blanco)

A method fusing spectral and image information with a one-dimensional convolutional neural network(1D-CNN) for the detection of moisture content in Orah mandarin (Citrus reticulata Blanco) was proposed. The 1D-CNN model integrated with three different attention modules (SEAM, ECAM, CBAM) and machine learning models were applied to individual spectrum and fused information by passing the traditional feature extraction stage. Additionally, the dimensionality reduction of hyperspectral images and extraction of one-dimensional color and textural features from the reduced images were performed, thus avoiding the large parameter volumes and efficiency decline inherent in the direct modeling of two-dimensional images. The results indicated that the 1D-CNN model with integrated attention modules exhibited clear advantages over machine learning models in handling multi-source information. The optimal machine learning model was determined to be the random forest (RF) model under the fusion information, with a correlation coefficient (R) of 0.8770 and a root mean square error (RMSE) of 0.0188 on the prediction set. The CBAM-1D-CNN model under the fusion information exhibited the best performance, with an R of 0.9172 and an RMSE of 0.0149 on the prediction set. The 1D-CNN models utilizing fusion information exhibited superior performance compared to single spectrum, and 1D-CNN with the fused information based on SEAM, ECAM, and CBAM respectively improved Rp by 4.54%, 0.18%, and 10.19% compared to the spectrum, with the RMSEP decreased by 11.70%, 14.06%, and 31.02%, respectively. The proposed approach of 1D-CNN integrated attention can obtain excellent regression results by only using one-dimensional data and without feature pre-extracting, reducing the complexity of the models, simplifying the calculation process, and rendering it a promising practical application.

Scene‐content‐sensitive real‐time adaptive foveated rendering

AbstractIn recent years, techniques for accelerating rendering by exploiting the limitations of the human visual system have become increasingly prevalent. The foveated rendering method significantly reduces the computational requirements during rendering by reducing image quality in peripheral regions. In this paper, we propose a scene‐content‐sensitive real‐time adaptive foveated rendering method. First, we pre‐render the three‐dimensional (3D) scene at a low resolution. Then, we utilize the low‐resolution pre‐rendered image as input to extract edge, local contrast, and color features. Subsequently, we generate a screen‐space region division map based on the gaze point position. Next, we calculate the visual importance of each 16 × 16 pixel tile based on edge, local contrast, color, and screen‐space region. We then map the visual importance to the shading rate to generate a shading rate control map for the current frame. Finally, we complete the rendering of the current frame based on variable rate shading technology. Experimental results demonstrate that our method effectively enhances the visual quality of images near the foveal region while generating high quality foveal region images. Furthermore, our method can significantly improve performance compared to per‐pixel shading method and existing scene‐content‐based foveated rendering methods.

Color Feature Extraction for Grape Variety Identification: Naïve Bayes Approach

The problem addressed in this research is the lack of an efficient and accurate method for automatically identifying grape varieties. Accurate identification is crucial for quality control in the agricultural and food industries, impacting product labeling, pricing, and consumer trust. The aim of this research is to develop an automated system to classify green, black, and red grapes using digital image processing technology. This research method employs Naïve Bayes classification combined with color feature extraction. Testing was conducted under two scenarios: a database scenario with predefined grape image datasets and an out-of-database scenario with images resembling grape colors. Image processing includes resizing images to 200x200 pixels, Gamma Correction, Gaussian filtering, conversion to Lab* color space, K-Means Clustering for segmentation, followed by feature extraction and Naïve Bayes classification. The results of this research are that in the database scenario, the system achieved accuracies of 98.33% with an 80:20 data split and 98.89% with a 70:30 split. In the out-of-database scenario, accuracies were 96.67% with an 80:20 split and 97.78% with a 70:30 split. The conclusion of this research is the proposed method provides a reliable and efficient solution for automatic grape variety identification, benefiting quality control in agriculture and food industries.

Analisis Kinerja Logistic Regression Classifier Berdasarkan Seleksi Fitur Warna, GLCM (Gray Level Co-occurrence Matrix) dan Bentuk (Studi Kasus Jenis Ketupat Khas Bali)

Ketupat is a unique culture and tradition in Bali. Ketupat is often used as a banten offering. This unique procession of the ketupat war is held once a year to coincide with the Purnama Kapat. The ketupat war is a traditional event with participants throwing ketupat at each other. It aims to be grateful for all the gifts that the Creator has given to humans in this world. This tradition has existed since the 1970s with the beginning of its appearance involving two shirtless men. The ketupat war ceremony is part of the yadnya as the basis for the return of the Tri Rna. In today's era of the development of science and technology, people are only focused on meeting material needs, understanding the meaning of rituals and religious events is decreasing, and it is considered an incriminating or even meaningless activity. In this study, a dataset of 90 was used with 18 types of ketupat used, namely: bagia, batu dam-pulan, bracelet, kale, mattress, kedis, kepel, kroso, lepet, pengambean, sai, cow, sari, sirikan, suna, tulud, tumpeng. This study aims to determine the per-formance of Logistic Regression Classifier by using color, HSV, GLCM and shape features. The dataset used is a typical Balinese ketupat type of 90 data. In this test, 4 test scenarios are used: 1) The first test scenario, the dataset input performs the preprocessing process, HSV imagery, and the color feature extraction process where the output results are in the form of hue, saturation and value values in the form of excel files. 2) The second test scenario, the input dataset performs a preprocessing process, grayscale, and performs the GLCM feature extraction process where the output results are in the form of feature values of angle 00, angle 450, angle 900 and angle 1350 in the form of an excel file. 3) The third test scenario, the dataset input performs the preprocessing process, binner, performs the form feature extraction process where the output results are in the form of metric, eccentricity in the form of excel files. 4) The fourth test scenario, from the 3 (three) feature extractions carried out, the new dataset, the next stage is to implement the Logistic Re-gression classifier method to obtain accuracy values. Based on the results of the classification, testing and analysis, the level of accuracy from each of them was obtained with a training accuracy value of 69.84% and a testing accuracy of 22.2%, which means that the classification method was declared ineffective in analyzing the features used in the ketupat classification, so it is necessary to compare it with other methods so that it gets an accuracy result above 90%.

Integrating Image Analysis and Machine Learning for Moisture Prediction and Appearance Quality Evaluation: A Case Study of Kiwifruit Drying Pretreatment.

The appearance of dried fruit clearly influences the consumer's perception of the quality of the product but is a subtle and nuanced characteristic that is difficult to quantitatively measure, especially online. This paper describes a method that combines several simple strategies to assess a suitable surrogate for the elusive quality using imaging, combined with multivariate statistics and machine learning. With such a convenient tool, this study also shows how one can vary the pretreatments and drying conditions to optimize the resultant product quality. Specifically, an image batch processing method was developed to extract color (hue, saturation, and value) and morphological (area, perimeter, and compactness) features. The accuracy of this method was verified using data from a case study experiment on the pretreatment of hot-air-dried kiwifruit slices. Based on the extracted image features, partial least squares and random forest models were developed to satisfactorily predict the moisture ratio (MR) during drying process. The MR of kiwifruit slices during drying could be accurately predicted from changes in appearance without using any weighing device. This study also explored determining the optimal drying strategy based on appearance quality using principal component analysis. Optimal drying was achieved at 60 °C with 4 mm thick slices under ultrasonic pretreatment. For the 70 °C, 6 mm sample groups, citric acid showed decent performance.

Remote sensing image dehazing using generative adversarial network with texture and color space enhancement

Remote sensing is gradually playing an important role in the detection of ground information. However, the quality of remote-sensing images has always suffered from unexpected natural conditions, such as intense haze phenomenon. Recently, convolutional neural networks (CNNs) have been applied to deal with dehazing problems, and some important findings have been obtained. Unfortunately, the performance of these classical CNN-based methods still needs further enhancement owing to their limited feature extraction capability. As a critical branch of CNNs, the generative adversarial network (GAN), composed of a generator and discriminator, has become a hot research topic and is considered a feasible approach to solving the dehazing problems. In this study, a novel dehazed generative adversarial network (GAN) is proposed to reconstruct the clean images from the hazy ones. For the generator network of the proposed GAN, the color and luminance feature extraction module and the high-frequency feature extraction module aim to extract multi-scale features and color space characteristics, which help the network to acquire texture, color, and luminance information. Meanwhile, a color loss function based on hue saturation value (HSV) is also proposed to enhance the performance in color recovery. For the discriminator network, a parallel structure is designed to enhance the extraction of texture and background information. Synthetic and real hazy images are used to check the performance of the proposed method. The experimental results demonstrate that the performance can significantly improve the image quality with a significant increment in peak-signal-to-noise ratio (PSNR). Compared with other popular methods, the dehazing results of the proposed method closely resemble haze-free images.

A superpixel‐based color difference detection algorithm for dyed fabrics

AbstractIn order to enhance the accuracy and efficiency of color difference detection for dyed fabrics, this article introduces a novel superpixel‐based color difference detection algorithm, which exhibits three significant improvements over the previous algorithm. First, it proposes a more rational color feature extraction method to extract more detailed color features from fabric images, which uses superpixels as basic units for both color feature extraction and color difference detection. Second, the algorithm employs the CIEDE2000 color difference formula to calculate color difference values. This eliminates the need for obtaining standard fabric images as references in advance, making the process more convenient and practical. Finally, an area‐based color difference calculation method is proposed based on the human visual mechanism, ensuring that the evaluation results are more consistent with human eye observations. Experimental results demonstrate that the proposed algorithm provides more accurate and visually consistent color difference assessments compared to the previous algorithm.

DON6D: a decoupled one-stage network for 6D pose estimation

The six-dimensional (6D) pose object estimation is a key task in robotic manipulation and grasping scenes. Many existing two-stage solutions with a slow inference speed require extra refinement to handle the challenges of variations in lighting, sensor noise, object occlusion, and truncation. To address these challenges, this work proposes a decoupled one-stage network (DON6D) model for 6D pose estimation that improves inference speed on the premise of maintaining accuracy. Particularly, since the RGB images are aligned with the RGB-D images, the proposed DON6D first uses a two-dimensional detection network to locate the interested objects in RGB-D images. Then, a module of feature extraction and fusion is used to extract color and geometric features fully. Further, dual data augmentation is performed to enhance the generalization ability of the proposed model. Finally, the features are fused, and an attention residual encoder–decoder, which can improve the pose estimation performance to obtain an accurate 6D pose, is introduced. The proposed DON6D model is evaluated on the LINEMOD and YCB-Video datasets. The results demonstrate that the proposed DON6D is superior to several state-of-the-art methods regarding the ADD(-S) and ADD(-S) AUC metrics.

Optimal Artificial Neural Network-based Fabric Defect Detection and Classification

Automated Fabric Defect (FD) detection plays a crucial role in industrial automation within fabric production. Traditionally, the identification of FDs heavily relies on manual assessment, facilitating prompt repairs of minor defects. However, the efficiency of manual recognition diminishes significantly as labor working hours increase. Consequently, there is a pressing need to introduce an automated analysis method for FD recognition to reduce labor costs, minimize errors, and improve fabric quality. Many researchers have devised defect detection systems utilizing Machine Learning (ML) approaches, enabling swift, accurate, and efficient identification of defects. This study presents the Optimal Artificial Neural Network-based Fabric Defect Detection and Classification (OANN-FDDC) technique. The OANN-FDDC technique exploits handcrafted features with a parameter-tuning strategy for effectively detecting the FD process. To obtain this, the OANN-FDDC technique employs CLAHE and Bilateral Filtering (BF) model-based contrast augmentation and noise removal. Besides, the OANN-FDDC technique extracts shape, texture, and color features. For FD detection, the ANN method is utilized. To improve the detection results of the ANN method, the Root Mean Square Propagation (RMSProp) optimization technique is used for the parameter selection process. The simulation outputs of the OANN-FDDC technique were examined on an open fabric image database. The experimental results of the OANN-FDDC technique implied a better outcome than the 96.97% of other recent approaches.