Empirical Examination of Color Spaces in Deep Convolution Networks

In this paper we present an empirical examination of deep convolution neural network (DCNN) performance in different color spaces for the classical problem of image recognition/classification. Most such deep learning architectures or networks are applied on RGB color space image data set, so our objective is to study DCNNs performance in other color spaces. We describe the design of our novel experiment and present results on whether deep learning networks for image recognition task is invariant to color spaces or not. In this study, we have analyzed the performance of 3 popular DCNNs (VGGNet, ResNet, GoogleNet) by providing input images in 5 different color spaces(RGB, normalized RGB, YCbCr, HSV , CIE-Lab) and compared performance in terms of test accuracy, test loss, and validation loss. All these combination of networks and color spaces are investigated on two datasets- CIFAR 10 and LINNAEUS 5. Our experimental results show that CNNs are variant to color spaces as different color spaces have different performance results for image classification task.

Download Full-text

Research on the Application of Artificial Intelligence Machine Learning Technology in Improving the Accuracy of Engineering Image Processing

Journal of Physics Conference Series ◽

10.1088/1742-6596/2083/4/042007 ◽

2021 ◽

Vol 2083 (4) ◽

pp. 042007

Author(s):

Xiaowen Liu ◽

Juncheng Lei

Keyword(s):

Machine Learning ◽

Feature Extraction ◽

Deep Learning ◽

Image Recognition ◽

Recognition Performance ◽

Color Space ◽

Gaussian Model ◽

Image Feature ◽

Layer By Layer ◽

Image Information

Abstract Image recognition technology mainly includes image feature extraction and classification recognition. Feature extraction is the key link, which determines whether the recognition performance is good or bad. Deep learning builds a model by building a hierarchical model structure like the human brain, extracting features layer by layer from the data. Applying deep learning to image recognition can further improve the accuracy of image recognition. Based on the idea of clustering, this article establishes a multi-mix Gaussian model for engineering image information in RGB color space through offline learning and expectation-maximization algorithms, to obtain a multi-mix cluster representation of engineering image information. Then use the sparse Gaussian machine learning model on the YCrCb color space to quickly learn the distribution of engineering images online, and design an engineering image recognizer based on multi-color space information.

Download Full-text

Estimation of Fractal Dimension in Different Color Model

International Journal of Knowledge Discovery in Bioinformatics ◽

10.4018/ijkdb.2018010106 ◽

2018 ◽

Vol 8 (1) ◽

pp. 75-93

Author(s):

Sumitra Kisan ◽

Sarojananda Mishra ◽

Ajay Chawda ◽

Sanjay Nayak

Keyword(s):

Fractal Dimension ◽

Color Space ◽

Vital Role ◽

Color Images ◽

Color Spaces ◽

Fractal Surfaces ◽

Test Operation ◽

Novel Method ◽

Rgb Color Space ◽

Different Color

This article describes how the term fractal dimension (FD) plays a vital role in fractal geometry. It is a degree that distinguishes the complexity and the irregularity of fractals, denoting the amount of space filled up. There are many procedures to evaluate the dimension for fractal surfaces, like box count, differential box count, and the improved differential box count method. These methods are basically used for grey scale images. The authors' objective in this article is to estimate the fractal dimension of color images using different color models. The authors have proposed a novel method for the estimation in CMY and HSV color spaces. In order to achieve the result, they performed test operation by taking number of color images in RGB color space. The authors have presented their experimental results and discussed the issues that characterize the approach. At the end, the authors have concluded the article with the analysis of calculated FDs for images with different color space.

Download Full-text

Color Image Segmentation Based on Different Color Space Models Using Automatic GrabCut

The Scientific World JOURNAL ◽

10.1155/2014/126025 ◽

2014 ◽

Vol 2014 ◽

pp. 1-10 ◽

Cited By ~ 14

Author(s):

Dina Khattab ◽

Hala Mousher Ebied ◽

Ashraf Saad Hussein ◽

Mohamed Fahmy Tolba

Keyword(s):

Image Segmentation ◽

Comparative Study ◽

Color Image ◽

Color Space ◽

User Interaction ◽

Color Image Segmentation ◽

Space Representation ◽

Color Spaces ◽

Different Color ◽

The Comparative Study

This paper presents a comparative study using different color spaces to evaluate the performance of color image segmentation using the automatic GrabCut technique. GrabCut is considered as one of the semiautomatic image segmentation techniques, since it requires user interaction for the initialization of the segmentation process. The automation of the GrabCut technique is proposed as a modification of the original semiautomatic one in order to eliminate the user interaction. The automatic GrabCut utilizes the unsupervised Orchard and Bouman clustering technique for the initialization phase. Comparisons with the original GrabCut show the efficiency of the proposed automatic technique in terms of segmentation, quality, and accuracy. As no explicit color space is recommended for every segmentation problem, automatic GrabCut is applied withRGB,HSV,CMY,XYZ, andYUVcolor spaces. The comparative study and experimental results using different color images show thatRGBcolor space is the best color space representation for the set of the images used.

Download Full-text

Improving Color Space Conversion for Camera-Captured Images via Wide-Gamut Metadata

Color and Imaging Conference ◽

10.2352/issn.2169-2629.2020.28.30 ◽

2020 ◽

Vol 2020 (28) ◽

pp. 193-198

Author(s):

Hoang Le ◽

Mahmoud Afifi ◽

Michael S. Brown

Keyword(s):

Color Space ◽

Mapping Function ◽

Mapping Method ◽

Color Spaces ◽

Color Gamut ◽

Color Mapping ◽

Color Space Conversion ◽

Different Color ◽

Image Format ◽

Image State

Color space conversion is the process of converting color values in an image from one color space to another. Color space conversion is challenging because different color spaces have different sized gamuts. For example, when converting an image encoded in a medium-sized color gamut (e.g., AdobeRGB or Display-P3) to a small color gamut (e.g., sRGB), color values may need to be compressed in a many-to-one manner (i.e., multiple colors in the source gamut will map to a single color in the target gamut). If we try to convert this sRGB-encoded image back to a wider gamut color encoding, it can be challenging to recover the original colors due to the color fidelity loss. We propose a method to address this problem by embedding wide-gamut metadata inside saved images captured by a camera. Our key insight is that in the camera hardware, a captured image is converted to an intermediate wide-gamut color space (i.e., ProPhoto) as part of the processing pipeline. This wide-gamut image representation is then saved to a display color space and saved in an image format such as JPEG or HEIC. Our method proposes to include a small sub-sampling of the color values from the ProPhoto image state in the camera to the final saved JPEG/HEIC image. We demonstrate that having this additional wide-gamut metadata available during color space conversion greatly assists in constructing a color mapping function to convert between color spaces. Our experiments show our metadata-assisted color mapping method provides a notable improvement (up to 60% in terms of E) over conventional color space methods using perceptual rendering intent. In addition, we show how to extend our approach to perform adaptive color space conversion based spatially over the image for additional improvements.

Download Full-text

Impacts of Color Space Transformations on Dysplastic Nuclei Segmentation Using CNN

10.5753/wvc.2020.13475 ◽

2020 ◽

Author(s):

Dalí Dos Santos ◽

Adriano Silva ◽

Paulo De Faria ◽

Bruno Travençolo ◽

Marcelo Do Nascimento

Keyword(s):

Color Space ◽

Precancerous Lesion ◽

Epithelial Dysplasia ◽

Color Spaces ◽

Oral Epithelial Dysplasia ◽

Nuclei Segmentation ◽

Space Transformation ◽

Different Color ◽

Oral Epithelial ◽

The Impact

Oral epithelial dysplasia is a common precancerous lesion type that can be graded as mild, moderate and severe. Although not all oral epithelial dysplasia become cancer over time, this premalignant condition has a significant rate of progressing to cancer and the early treatment has been shown to be considerably more successful. The diagnosis and distinctions between mild, moderate, and severe grades are made by pathologists through a complex and time-consuming process where some cytological features, including nuclear shape, are analysed. The use of computer-aided diagnosis can be applied as a tool to aid and enhance the pathologist decisions. Recently, deep learning based methods are earning more and more attention and have been successfully applied to nuclei segmentation problems in several scenarios. In this paper, we evaluated the impact of different color spaces transformations for automated nuclei segmentation on histological images of oral dysplastic tissues using fully convolutional neural networks (CNN). The CNN were trained using different color spaces from a dataset of tongue images from mice diagnosed with oral epithelial dysplasia. The CIE L*a*b* color space transformation achieved the best averaged accuracy over all analyzed color space configurations (88.2%). The results show that the chrominance information, or the color values, does not play the most significant role for nuclei segmentation purpose on a mice tongue histopathological images dataset.

Download Full-text

Evaluasi Performasi Ruang Warna pada Klasifikasi Diabetic Retinophaty Menggunakan Convolution Neural Network

Jurnal Teknologi Informasi dan Ilmu Komputer ◽

10.25126/jtiik.2021834459 ◽

2021 ◽

Vol 8 (3) ◽

pp. 619

Author(s):

Candra Dewi ◽

Andri Santoso ◽

Indriati Indriati ◽

Nadia Artha Dewi ◽

Yoke Kusuma Arbawa

Keyword(s):

Neural Network ◽

Diabetic Retinopathy ◽

Deep Learning ◽

Convolutional Neural Network ◽

Learning Algorithm ◽

Color Space ◽

Training Data ◽

Color Spaces ◽

Deep Learning Algorithm ◽

Optimal Accuracy

Semakin meningkatnya jumlah penderita diabetes menjadi salah satu faktor penyebab semakin tingginya penderita penyakit diabetic retinophaty. Salah satu citra yang digunakan oleh dokter mata untuk mengidentifikasi diabetic retinophaty adalah foto retina. Dalam penelitian ini dilakukan pengenalan penyakit diabetic retinophaty secara otomatis menggunakan citra fundus retina dan algoritme Convolutional Neural Network (CNN) yang merupakan variasi dari algoritme Deep Learning. Kendala yang ditemukan dalam proses pengenalan adalah warna retina yang cenderung merah kekuningan sehingga ruang warna RGB tidak menghasilkan akurasi yang optimal. Oleh karena itu, dalam penelitian ini dilakukan pengujian pada berbagai ruang warna untuk mendapatkan hasil yang lebih baik. Dari hasil uji coba menggunakan 1000 data pada ruang warna RGB, HSI, YUV dan L*a*b* memberikan hasil yang kurang optimal pada data seimbang dimana akurasi terbaik masih dibawah 50%. Namun pada data tidak seimbang menghasilkan akurasi yang cukup tinggi yaitu 83,53% pada ruang warna YUV dengan pengujian pada data latih dan akurasi 74,40% dengan data uji pada semua ruang warna. AbstractIncreasing the number of people with diabetes is one of the factors causing the high number of people with diabetic retinopathy. One of the images used by ophthalmologists to identify diabetic retinopathy is a retinal photo. In this research, the identification of diabetic retinopathy is done automatically using retinal fundus images and the Convolutional Neural Network (CNN) algorithm, which is a variation of the Deep Learning algorithm. The obstacle found in the recognition process is the color of the retina which tends to be yellowish red so that the RGB color space does not produce optimal accuracy. Therefore, in this research, various color spaces were tested to get better results. From the results of trials using 1000 images data in the color space of RGB, HSI, YUV and L * a * b * give suboptimal results on balanced data where the best accuracy is still below 50%. However, the unbalanced data gives a fairly high accuracy of 83.53% with training data on the YUV color space and 74,40% with testing data on all color spaces.

Download Full-text

RGB Image Color Space Transformations

The Insight Journal ◽

10.54294/sn4ute ◽

2010 ◽

Author(s):

Robert Tamburo

Keyword(s):

Source Code ◽

Color Space ◽

Color Spaces ◽

Different Color ◽

Rgb Image

This paper describes a set of pixel accessors that transform RGB pixel values to a different color space. Accessors for the HSI, XYZ, Yuv, YUV, HSV, Lab, Luv, HSL, CMY, and CMYK color spaces are provided here. This paper is accompanied with source code for the pixel accessors and test, test images and parameters, and expected output images.Note: Set() methods are incorrect. Will provide revision by 12.17.2010.

Download Full-text

Multifocus Color Image Fusion Based on NSST and PCNN

Journal of Sensors ◽

10.1155/2016/8359602 ◽

2016 ◽

Vol 2016 ◽

pp. 1-12 ◽

Cited By ~ 31

Author(s):

Xin Jin ◽

Rencan Nie ◽

Dongming Zhou ◽

Quan Wang ◽

Kangjian He

Keyword(s):

Neural Networks ◽

Image Fusion ◽

Color Image ◽

Color Space ◽

Color Spaces ◽

Fusion Algorithm ◽

Hsv Color Space ◽

Coupled Neural Networks ◽

Different Color ◽

Pulse Coupled Neural Networks

This paper proposed an effective multifocus color image fusion algorithm based on nonsubsampled shearlet transform (NSST) and pulse coupled neural networks (PCNN); the algorithm can be used in different color spaces. In this paper, we take HSV color space as an example, H component is clustered by adaptive simplified PCNN (S-PCNN), and then the H component is fused according to oscillation frequency graph (OFG) of S-PCNN; at the same time, S and V components are decomposed by NSST, and different fusion rules are utilized to fuse the obtained results. Finally, inverse HSV transform is performed to get the RGB color image. The experimental results indicate that the proposed color image fusion algorithm is more efficient than other common color image fusion algorithms.

Download Full-text

Wheel Hub Defects Image Recognition Base on Zero-Shot Learning

Applied Sciences ◽

10.3390/app11041529 ◽

2021 ◽

Vol 11 (4) ◽

pp. 1529

Author(s):

Xiaohong Sun ◽

Jinan Gu ◽

Meimei Wang ◽

Yanhua Meng ◽

Huichao Shi

Keyword(s):

Deep Learning ◽

Image Recognition ◽

Large Scale ◽

Domain Adaptation ◽

Data Sets ◽

Human Beings ◽

Data Set ◽

Defect Image ◽

Product Surface ◽

Projection Domain

In the wheel hub industry, the quality control of the product surface determines the subsequent processing, which can be realized through the hub defect image recognition based on deep learning. Although the existing methods based on deep learning have reached the level of human beings, they rely on large-scale training sets, however, these models are completely unable to cope with the situation without samples. Therefore, in this paper, a generalized zero-shot learning framework for hub defect image recognition was built. First, a reverse mapping strategy was adopted to reduce the hubness problem, then a domain adaptation measure was employed to alleviate the projection domain shift problem, and finally, a scaling calibration strategy was used to avoid the recognition preference of seen defects. The proposed model was validated using two data sets, VOC2007 and the self-built hub defect data set, and the results showed that the method performed better than the current popular methods.

Download Full-text

Deep Convolutional Neural Network for Ulcer Recognition in Wireless Capsule Endoscopy: Experimental Feasibility and Optimization

Computational and Mathematical Methods in Medicine ◽

10.1155/2019/7546215 ◽

2019 ◽

Vol 2019 ◽

pp. 1-14 ◽

Cited By ~ 4

Author(s):

Sen Wang ◽

Yuxiang Xing ◽

Li Zhang ◽

Hewei Gao ◽

Hao Zhang

Keyword(s):

Deep Learning ◽

Capsule Endoscopy ◽

Recognition Task ◽

Wireless Capsule Endoscopy ◽

Test Accuracy ◽

Learning Networks ◽

Deep Convolutional Neural Networks ◽

Deep Cnn ◽

Wireless Capsule ◽

Better Than

Wireless capsule endoscopy (WCE) has developed rapidly over the last several years and now enables physicians to examine the gastrointestinal tract without surgical operation. However, a large number of images must be analyzed to obtain a diagnosis. Deep convolutional neural networks (CNNs) have demonstrated impressive performance in different computer vision tasks. Thus, in this work, we aim to explore the feasibility of deep learning for ulcer recognition and optimize a CNN-based ulcer recognition architecture for WCE images. By analyzing the ulcer recognition task and characteristics of classic deep learning networks, we propose a HAnet architecture that uses ResNet-34 as the base network and fuses hyper features from the shallow layer with deep features in deeper layers to provide final diagnostic decisions. 1,416 independent WCE videos are collected for this study. The overall test accuracy of our HAnet is 92.05%, and its sensitivity and specificity are 91.64% and 92.42%, respectively. According to our comparisons of F1, F2, and ROC-AUC, the proposed method performs better than several off-the-shelf CNN models, including VGG, DenseNet, and Inception-ResNet-v2, and classical machine learning methods with handcrafted features for WCE image classification. Overall, this study demonstrates that recognizing ulcers in WCE images via the deep CNN method is feasible and could help reduce the tedious image reading work of physicians. Moreover, our HAnet architecture tailored for this problem gives a fine choice for the design of network structure.

Download Full-text