Color Vector Research Articles

In-field citrus detection and localisation based on RGB-D image analysis

In-field citrus detection and localisation are highly challenging tasks due to varying illumination conditions, partial occlusion of citrus, and the colour variation of citrus at different stages of maturity. A reliable algorithm based on red-green-blue-depth (RGB-D) images was developed to detect and locate citrus in real, outdoor orchard environments for robotic harvesting. A depth filter and a Bayes-classifier-based image segmentation method were first developed to exclude as many backgrounds as possible. A density clustering method was then used to group adjacent points in the filtered RGB-D images into clusters, where each cluster represents a possible citrus. A colour, gradient, and geometry feature-based support vector machine classifier was trained to remove false positives. To test the method, a dataset with 506 RGB-D images was acquired in a citrus orchard on sunny and cloudy days. Results showed that the proposed algorithm was robust with an F1 score of 0.9197; the positioning errors in the x, y and z directions were 7.0 ± 2.5 mm, −4.0 ± 3.0 mm and 13.0 ± 3.0 mm, respectively, and the sizing error was −1.0 ± 4.0 mm. These excellent performance values demonstrate that the proposed method could be used to guide a citrus-harvesting robot.

Removal of impulse noise in color images based on convolutional neural network

A new denoising framework based on deep convolutional neural network for suppressing impulse noise in color images is proposed in this paper. The proposed framework consists of two modules: noise detection and image reconstruction, both of which are implemented by a deep convolutional neural network. First, a noise classifier network is trained to detect random-valued impulse noise in a color image, which not only can detect the noisy color vector pixels but also can further identify the corrupted channels of each noisy color pixel. Then, a sparse clean color image is computed by replacing the values of noisy channels with 0 and keeping other noise-free channels unchanged. Finally, the sparse clean color image is fed to another denoiser network to reconstruct the denoised image. Experimental results show that the proposed denoiser outperforms other state-of-the-art methods clearly in both performance measure and visual evaluation.

LRT Digest 3: New colour metrics and their use

Executive summary The two metrics traditionally used to quantify the colour properties of light sources are the correlated colour temperature and the CIE General Colour Rendering Index. With the arrival of LEDs as a major light source questions began to be asked about the merits of both of these metrics. The question asked about correlated colour temperature was how far should the chromaticity of a light source be allowed to depart from the Planckian locus before the light emitted could no longer be said to be white? A tolerance to such a departure ( Duv) already existed but now gathered much more attention. The questions asked about the CIE General Colour Rendering Index were more searching. The limitations of the CIE General Colour Rendering Index were explored and, as a result, several alternative approaches to quantifying the colour rendition properties of light sources were proposed. The most comprehensive approach was produced by the Illuminating Engineering Society of North America, first in its Technical Memorandum IES TM-30-15 and more recently in its revision, Technical Memorandum ANSI/IES TM-30-18, which has been accepted as an American Standard. Both these documents describe a system that contains two high-level summary metrics: One for the average fidelity, i.e. how accurately a test light source renders 99 colour samples relative to how they are rendered under a reference illuminant, and the other for the average colourfulness, i.e. the overall increase or decrease in colourfulness of the same colour samples under the same test light source compared to the same reference illuminant. Associated with these overall average metrics are a number of more detailed metrics and graphical presentations. These aim to quantify and illustrate the variations in fidelity and the direction and magnitude of the shifts in chroma and hue around the hue circle. Compared to the CIE General Colour Rendering Index or the CIE Fidelity Index, a metric published by the CIE in 2017, ANSI/IES TM-30-18 provides a more comprehensive approach to quantifying and understanding the effects of light source spectrum on the perception of colour. Unfortunately, the new colour metrics described in IES TM-30-15 and ANSI/IES TM-30-18 have not yet been accepted by the CIE. Despite this, some light source manufacturers have started to provide information on their products expressed in terms of the ANSI/IES TM-30-18 colour metrics and designers are beginning to request them. The expectation is that, eventually, the ANSI/IES TM-30-18 metrics will be adopted by many countries and authorities, because they provide a much more comprehensive description of the colour properties of a light source than the CIE General Colour Rendering Index. This will be of value to light source manufacturers and lighting designers as well as those who prepare lighting codes and guides. It is expected that in the future the minimum set of data considered acceptable for describing the colour properties of a light source are likely to be the correlated colour temperature and the associated Duv value, the CIE Fidelity Index or the matching ANSI/IES TM-30-18 Fidelity Index, together with the ANSI/IES TM-30-18 Gamut Index and the ANSI/IES TM-30-18 Colour Vector Graphic.

INDICADORES INDIRECTOS DE CONTAMINACIÓN RESIDUAL EN SUELOS Y SEDIMENTOS DE LA CUENCA DEL RÍO SONORA, MÉXICO

In Cananea, Sonora, the spill of 40 000 m 3 of an acid solution rich in iron and copper reached the main water streams of the Sonora river basin, affecting soils and sediments and giving them an anomalous yellowish brown coloration. In the present study, residual contamination (RC) of soils and sediments in the basin was monitored, evaluating color, pH and electrical conductivity (EC) in 41 samples obtained throughout the basin at various depths in the vertical and horizontal directions in a soil profile (TIIZ1B) affected by RC. These properties were supposed to be indirect indicators of RC because of their modification by the influence of the poured solution. Additionally, total concentrations of potentially toxic elements present in the poured solution were determined in soils and sediments. There is an acceptable lineal correlation (0.75) between EC and the yellow color vector (b*), and between these and the concentrations of As, Cu, Fe and Zn (r = 0.52-0.91). The TIIZ1B soil profile analysis allowed relating color, pH and EC with mineralogical changes and increase in the concentrations of As, Cu and Fe. The increase in color vectors to red (a*) and yellow tones is related to the rise in the Fe concentration and to the increase of brightness (L*), the latter being linked to the precipitation of calcium sulfate. Concentration of As and Fe increases in the portion of the soil with the most acidic pH values, while Cu increases with the increase of pH and a negative point of zero charge. Color, pH and EC are useful parameters in the indirect determination of RC.

New Hardware Architecture for Self-Organizing Map Used for Color Vector Quantization

In this paper, we present a new generic architectural approach of a Self-Organizing Map (SOM). The proposed architecture, called the Diagonal-SOM (D-SOM), is described as an Hardware–Description-Language as an intellectual property kernel with easily adjustable parameters.The D-SOM architecture is based on a generic formalism that exploits two levels of the nested parallelism of neurons and connections. This solution is therefore considered as a system based on the cooperation of a distributed set of independent computations. The organization and structure of these calculations process an oriented data flow in order to find a better treatment distribution between different neuroprocessors. To validate the D-SOM architecture, we evaluate the performance of several SOM network architectures after their integration on a Xilinx Virtex-7 Field Programmable Gate Array support. The proposed solution allows the easy adaptation of learning to a large number of SOM topologies without any considerable design effort. [Formula: see text] SOM hardware is validated through FPGA implementation, where temporal performance is almost twice as fast as that obtained in the recent literature. The suggested D-SOM architecture is also validated through simulation on variable-sized SOM networks applied to color vector quantization.

光场相机精确色彩矢量约束下的超分辨率算法

光场相机精确色彩矢量约束下的超分辨率算法

Prediction of kiwifruit orchard characteristics from satellite images

The dry matter content of New Zealand kiwifruit is currently measured using destructive testing of a 90-fruit sample. Dry matter content varies from 14 to 20% with a standard deviation of 1.05% in the test measurement. This work investigates the use of multispectral data from satellite images of kiwifruit orchards to predict the dry matter of both green and gold kiwifruit. A novel method is developed that reduces the four-dimensional satellite data to three-dimensional unit color vectors and these show a strong linear relationship with measured dry matter. Regression on the data from a set of 20 ‘training’ orchards yielded predictions of dry matter with a standard deviation of 0.76%. When the resulting model was applied to nine test orchards the standard deviation of predicted dry matter was 0.73%. The prediction of dry matter was accurate even when applied to images taken when the fruit was too immature for the standard dry matter measurement test. Therefore, satellite image data may provide a more accurate and non-destructive alternative to the standard 90-fruit test method. It can also provide a way to visualize the variation in dry matter content of the fruit in an orchard; showing regions that would benefit from remedial action and defining areas where fruit harvesting is optimal. Several vegetation indices, derived from satellite image data, are reported in the literature. The strength of the linear correlation between dry matter and twelve common vegetation indices was tested but was found to be much weaker than the correlation developed in this research.

Novel chromaticity similarity based color texture descriptor for digital pathology image analysis.

Pathology images are color in nature due to the use of chemical staining in biopsy examination. Aware of the high color diagnosticity in pathology images, this work introduces a compact rotation-invariant texture descriptor, named quantized diagnostic counter-color pattern (QDCP), for digital pathology image understanding. On the basis of color similarity quantified by the inner product of unit-length color vectors, local counter-color textons are indexed first. Then the underlined distribution of QDCP indexes is estimated by an image-wise histogram. Since QDCP is computed based on color difference directly, it is robust to small color variation usually observed in pathology images. This study also discusses QDCP’s extraction, parameter settings, and feature fusion techniques in a generic pathology image analysis pipeline, and introduces two more descriptors QDCP-LBP and QDCP/LBP. Experimentation on public pathology image sets suggests that the introduced color texture descriptors, especially QDCP-LBP, outperform prior color texture features in terms of strong descriptive power, low computational complexity, and high adaptability to different image sets.

Construction Feature of Intrinsic Color Vector Span Space on Appearance of Tibetan Traditional Architecture from Digital Images

This article is focused on the qualitative analysis in the color art analysis of Tibetan traditional architecture(TTA) while lack of quantitative analysis, the definitions of visual distinguishable function, color vector distinguishable level and intrinsic color vector span distinguishable space(ICVSS) are introduced firstly, method for ICVSS is put forward by these definitions and simple set operation about TTA, which is used as the main manifestation of the color features of the TTA. Finally, numerical tests are implemented with the digital images and color point cloud data set is inputted by Matlab programming. The experiments are proved that the method is simple, accurate and effective, and agree with the works of the existing qualitative analysis.

Fuzzy averaging filter for impulse noise reduction in colour images with a correction step

In this paper we propose a fuzzy detection and reduction method for impulse noise in colour images. Detection is based on the fuzzyfication of a well-known statistic called ROD. The noise degrees obtained are used to reduce impulses by employing a fuzzy averaging between the input colour vector and a robust estimate of noise-free colour vector within the input neighbourhood. Fuzzy averaging has some advantages in terms of both noise reduction and detail preservation in front of detect and replace approaches because of threshold based decisions of the latter. However, robustness of the former is lower. We solve this problem by including a correction mechanism that checks the fuzzy noise degree of the output and replaces it with a robust colour vector either when noise has not been properly reduced or when a colour artefact has been introduced. We carry out a thorough study of the method parameter setting and give a convenient and robust setting. Experimental results show that our approach is very robust in front of four different types of impulse noise.

An improved Mahalanobis distance-based colour segmentation method for rural building recognition

Aiming at the rapid identification of rural buildings in complex environments from high-spatialresolution images, an improved Mahalanobis distance colour segmentation method (IMDCSM) is proposed and realised in Red, Green and Blue (RGB) space. Vector sets of a lower discrete degree are obtained by filtering the colour vector sets of the building samples, and a standard ellipsoid equation can be constructed based on these vector sets. The threshold of interested colour range can be flexibly and intuitively selected by changing the shape and size of this ellipsoid. Then, according to the relationship between the location of the image pixel colour vector and the ellipsoid, all building information can be extracted quickly. To verify the effectiveness of the proposed method, unmanned aerial vehicle (UAV) images of two areas in the suburbs of Chengdu city and Deyang city were utilised as experimental data for image segmentation, and the existing colour segmentation method based on the Mahalanobis distance was selected as an indicator to assess the effectiveness of this method. The experimental results demonstrate that the completeness and correctness of this method reached 95% and 83.0%, respectively, values that are higher than those of the Mahalanobis distance colour segmentation method (MDCSM). In general, this method is suitable for the rapid extraction of rural building information, and provides a new threshold selection method for classification.

Spatio-Temporal Disocclusion Filling Using Novel Sprite Cells

Depth image-based rendering is an important technique for virtual view synthesis with limited 3-D data. However, occluded areas result in disocclusions in synthesized images. Filling of the disocclusions in a plausible manner is a critical task in virtual view synthesis. In addition to spatial consistency, temporal consistency of the filled regions also affects the visual quality. In this paper, we propose a novel codebook method called sprite cell for filling the disocclusions with high spatial and temporal consistency. Each codeword consists of a color vector, depth value, frame log, and the confidence score of the corresponding pixel. In contrast with the existing methods that reuse the filling results of previous frames without considering their accuracy, the proposed method estimates the confidence scores of the filling results to prevent temporal continuation of filling errors. Moreover, we introduce a method to correct the luminance of filled disocclusions that compensates for the change of scenes. The experimental results show that the proposed method achieves both objective and subjective improvements over the state-of-the-art methods. The sequences synthesized with the proposed method have higher spatio-temporal consistency.

Image hashing with color vector angle

Color vector angle (CVA) is an important feature of processing color images and has been successfully developed and used in real applications, such as edge detection, indexing and retrieval of images. However, it is unsolved how to apply the CVA to efficiently generating an image hash. Also, most image hashing algorithms choose luminance component of color image for hash generation and cannot well capture the color information of images. To tackle these issues, we study efficient image hashing algorithms with the histogram of CVAs, called HCVA hashing. The histogram is first extracted from those angles that are in the biggest circle inscribed inside the normalized image. And then, it is compressed to make a short hash. We conducted some experiments to assess the performance, and illustrated that the DCT (Discrete Cosine Transform) is the best one of that cooperating with HCVA at generating hashes, as well as the HCVA hashing is robust and promising.

Models of colour quality over a wide range of spectral power distributions

Twenty-four real spectra were strategically structured to have systematic variation in average fidelity (IES Rf = 65, 75, 85, 95), average gamut (IES Rg = 80, 90, 100, 110, 120) and gamut shape (nominally saturating or desaturating red colour evaluation samples), at 3500 K. For each spectrum, 20 participants rated a set of objects along continuous scales of naturalness, preference, vividness and skin preference. Most of the top rated spectra, along all four scales, had average gamut greater than or equal to 100 (IES Rg ≥ 100) and did not desaturate red hues (IES Rcs,h16 ≥ 0%). Best fit models illustrate that average fidelity (i.e. IES Rf) and a proxy for red saturation (i.e. IES Rcs,h16, Rcs,h1) are salient derived measures. The results illustrate that a two-metric system of colour rendition, comprised of average metrics for fidelity and gamut, cannot fully describe colour quality and underscore the importance of a colour rendering graphic (i.e. IES Colour Vector Graphic). We propose and illustrate the utility of quantifying the shape of the IES TM-30-15 Color Vector Graphic with a best-fit ellipse.

Color and Vector Flow Imaging in Parallel Ultrasound With Sub-Nyquist Sampling.

RF acquisition with a high-performance multichannel ultrasound system generates massive data sets in short periods of time, especially in "ultrafast" ultrasound when digital receive beamforming is required. Sampling at a rate four times the carrier frequency is the standard procedure since this rule complies with the Nyquist-Shannon sampling theorem and simplifies quadrature sampling. Bandpass sampling (or undersampling) outputs a bandpass signal at a rate lower than the maximal frequency without harmful aliasing. Advantages over Nyquist sampling are reduced storage volumes and data workflow, and simplified digital signal processing tasks. We used RF undersampling in color flow imaging (CFI) and vector flow imaging (VFI) to decrease data volume significantly (factor of 3 to 13 in our configurations). CFI and VFI with Nyquist and sub-Nyquist samplings were compared in vitro and in vivo. The estimate errors due to undersampling were small or marginal, which illustrates that Doppler and vector Doppler images can be correctly computed with a drastically reduced amount of RF samples. Undersampling can be a method of choice in CFI and VFI to avoid information overload and reduce data transfer and storage.

Development of Urine Hydration System Based on Urine Color and Support Vector Machine

Nowadays, to maintain hydration in human body has becoming an important issue in health research. The water needs depend on many factors like body size, dietary intake, gender and physical activity, consequently the indicator of hydration status is highly individual. Based on recent research that said the urine color is reliable indicator for hydration status, we would like to develop a prototype for detecting hydration status automatically. To use the color indicator precisely, we tested several color sensors to performance in detecting color and chose TCS34725 as the color sensor. The prototype of urine hydration system (UHS) is designed to record hydration status data in daily basis in cloud computing and the urine information can be accessed by Android smartphone. The prototype employs a set of microcontrollers and sensors as IoT devices and support vector machine (SVM) as a classifier. To evaluate the accuracy of hydration status, we compared the prediction with urine specific gravity (USG) as golden standard. The accuracy of our UHS prototype can reach up to 84%, which is nearly similar with the accuracy of manual prediction using only human eyes.

Inverse conversion algorithm for an all-optical depth coloring camera.

Three-dimensional (3D) metrology has received a lot of attention from academic and industrial communities due to its broad applications, such as 3D contents, 3D printing, and autonomous driving. The all-optical depth coloring (AODC) camera has some benefits in computation load since it extracts depth information of an object fully optically. The AODC camera represents the depth of the object as a variation of wavelength, and spectroscopy is generally required to measure the wavelength. However, in the AODC camera, the color vector in RGB color space is convertible inversely into the wavelength after projection on the normalized rgb plane because the detected spectrum through the gating part has a narrow bandwidth as a result of the width of the slit in the projection part. In this paper, we propose an inverse conversion algorithm from RGB color to depth without spectroscopy. Experimental results are presented to confirm its feasibility. Also, some practical limitations are discussed, resulting from the nonlinearity of the response of the image sensor and the widths of the slits in the projection part and the gating part.

Recognition of vision-based activities of daily living using linear predictive coding of histogram of directional derivative

In this paper, we have introduced a novel approach for recognition of activities of daily living (ADL). These activities are the ones that the human beings perform in daily life. At the object level, we used computational color model for efficient object segmentation and tracking to handle dynamic background change in indoor environment. To make it computationally efficient, cosine of the angle between the expected image color vector and current image color vector is used. At feature level, we have presented a linear predictive coding of histogram of directional derivative as a spatio-temporal descriptor. Our proposed descriptor describes the local object shape and appearance within cuboids effectively and distinctively. A multiclass support vector machine has been used to classify the human activities. The proposed framework for recognition of indoor human activity has been extensively validated on the benchmark of ADL datasets, with a focus that this methodology is robust and attains more precise human activity recognition rate as compared to current methodologies available.

Color image sharpening based on local color statistics

This paper presents an effective color image sharpening method, which is based on local color statistics. First, the variance of a set of color samples is measured by a scalar that is computed based on the sum of distances of color vectors, whereas other studies usually treat a color variance as a 3D vector. This is because what a variance expresses is the degree of the deviation of the image (vector) signal from its mean, indicating that describing this degree of deviation by a scalar is reasonable. Then, the local scalar variance and mean vector are combined together to measure the change of color image signal from a pixel to its neighboring ones, and the polarity of the change is determined by the change of luminance. Finally, based on the measure of the change, an effective sharpening operator is developed. Experimental results show that the proposed method excellently sharpens different kinds of color images and at the same time preserves image chromaticity well, and outperforms other typical sharpening techniques in both objective assessment and visual evaluation.

More extreme equitable colorings of decompositions of [formula omitted] and [formula omitted

An H-decomposition of a graph G is a partition P of E(G) into blocks, each element of which induces a copy of H. An (s,p)-equitable block-coloring of the H-decomposition is a coloring of the blocks in P with exactly s colors such that each vertex u is incident with blocks colored with exactly p colors, the blocks containing u being shared out as evenly as possible among the p color classes. Early results in the literature consider the existence of (s,p)-equitable block-colorings of K3- and C4-decompositions of G∈{Kv,Kv−F}, focusing on finding the value of χp′(v), the smallest s for which such a coloring exists.More recently the structure of such colorings has been considered, defining the color vector V(E)=(c1(E),c2(E),…,cs(E)) of an (s,p)-equitable block-coloring E of G, arranged in non-decreasing order, where ci(E) is the number of vertices in G incident with a block of color i. The most interesting cases are where χp′(v)>p, finding the range of c1(E) and of cs(E) among all (s,p)-equitable block-colorings E of G being the main aim. The largest c1 can be and the smallest cs can be for C4-decompositions of Kv−F have been previously found. In this paper the value of the remaining two parameters of most interest are found; namely the smallest c1 can be and the largest cs can be.The extreme colorings found in the main results follow from another problem addressed in this paper: finding extreme (s,p)-equitable edge-colorings of Kv.An important facet of the paper is that the powerful proof technique of graph amalgamations is used for the first time to obtain (s,p)-equitable block-colorings.