Color Machine Vision System Research Articles

External Defects and Soil Deposits Identification on Potato Tubers using 2CCD Camera and Principal Component Images

Precise recognition of potato external defects and the ability to identify defects and non-defect areas are in demand. Common scab represents a significant issue that requires detection, yet identifying the extent of common scab infection remains challenging when using a standard RGB camera. In this research, a 2CCD camera system that could obtain a set of RGB and near-infrared images, which could enhance defect detection, has been used. Image segmentation strategies based on a single principal component image and the principal component pseudo-colored image have been proposed to identify external potato defects while excluding soil deposits on the potato surface, often recognized as defects by the normal color machine vision system. Performance metrics calculation results show relatively good results, with segmentation true accuracy around 64% for both methods. Principal component pseudo-colored images were able to discriminate defects area and soil deposits in a single image. The methods presented in this paper could be used as the basis to develop further classification and grading algorithms.

Maize and weed classification using color indices with support vector data description in outdoor fields

An automated method for maize and weed detection is very important to efficiently remove weeds and precisely calculate the quantity of maize. Color features were used in this study to investigate a simple maize-detection method using a color machine-vision system. Conventional image segmentation methods based on RGB values cannot separate maize from weeds because of the highly similar image RGB values of these plants. Thus, a post-processing algorithm was developed to distinguish maize from weeds after image preprocessing. Color indices were used to develop a classification model. The nine optimal features were selected by principal component analysis to reduce the effect of illumination. Finally, support vector data description was used as a classifier to differentiate maize from the mixes of different species of weeds. Pictures were taken by a commercial camera and used to verify the stability of the algorithm. Results show that the overall accuracy for three years is 90.19%, 92.36% and 93.87%, respectively. And the color indices used in this work were stable under various weather conditions and over time.

Classification of Single Cereal Grain Kernel Using Shape Parameters Based on Machine Vision

The study was conducted to identify three types of non-touching grain kernels using a colour machine vision system. Images of individual cereal grain kernels were acquired using an camera. Shape feature was extracted from binary and edge images of cereal grain kernels obtained by iamge processing for classification. A total of 13 shape feature parameters, including region area, perimeter, length, width, the maximum radius, the smallest radius etc, were extracted from each kernel to use as input to the Bayesian classifier. Experimental results showed that the Bayesian classifier gave better classification with a calssificaiton accuracy of 99.67% for indica type rice, followed by 98.67% and 78.33% for japonica rice and glutinous rice using training set, respectively. The classification system was developed with Bayesian classifier that achieved an overall recognition rate of 92.22% with training data set and furthermore, a classification accuracy of 90% for the testing data set.

소고기 육색 등급 자동 판정을 위한 기계시각 시스템의 칼라 보정 및 정량화

This study was conducted to evaluate beef using a color machine vision system. The machine vision system has an advantage to measure larger area than a colorimeter and also could measure other quality factors like distribution of fats. However, the machine vision measurement is affected by system components. To measure the beef color with the machine vision system, the effect of color balancing control was tested and calibration model was developed. Neural network for color calibration which learned reference color patches showed a high correlation with colorimeter in L*a*b* coordinates and had an adaptability at various measurement environments. The trained network showed a very high correlation with the colorimeter when measuring beef color.

Changes in Brown Eggshell Color As the Hen Ages

The color of eggshells from eggs laid by commercial-type Hy-Line brown hens 25 wk of age was studied over a period of 10 mo. Color measurements were made by a color machine vision system and were analyzed using a mixed model to calculate between and within hen variances and to investigate the effect of time on shell color. Hens laid eggs with lighter colored shells as the flock aged, as evidenced by the lightness (L*) values increasing in time. A decrease in pigmentation was associated with a decrease in the amount of redness (a*) in the eggshell. When L* and a* values were corrected for egg weight, the rate of change in the L* and a* values decreased, indicating that size of the egg was a major factor affecting the color of the eggshell. These findings quantified the observations that older hens lay lighter colored eggs due to an increase in egg size associated with no proportionate change in the quantity of pigment deposited over the shell surface. Using a 2-stage sampling analysis and the variances between and within hens, sample sizes required to estimate the color of eggshells within 5% of the true mean were calculated. Accordingly, 11 eggs would need to be collected from each of the 51 hens housed for a study of brown eggshell color using the L*, a*, and b* (yellowness) coordinates.

Detection of Defects in Rice Seeds Using Machine Vision

Three image-processing algorithms were developed to detect external defects of rice seeds such as germ, disease, and incompletely closed glumes. The rice seeds used for this study involved five varieties: Jinyou402, Shanyou10, Zhongyou207, Jiayou, and IIyou. Images of the samples with both black and white backgrounds were acquired with a color machine vision system. Each original image was preprocessed to create a mask for the seed region. For judging the presence of germ, 16 contour features were extracted and analyzed using principal components analysis. In addition to this, four back-propagation neural networks were created and trained with typical data sets of the four varieties. The algorithm developed for recognition of germ achieved an average accuracy of 99.4% for normal seeds and 91.9% for germinated seeds on panicle. The mean hue value and its deviation of the seed region determined with a block method were extracted as features of disease recognition. The corresponding algorithm developed for inspecting diseased seeds based on color features achieved an accuracy of 92.1% for normal seeds, 94.8% for spot-diseased seeds, and 91.1% for severely diseased seeds. Using radon transform, the group number of post-processing images proved to be a good indicator of incompletely closed glumes. The relevant algorithm was developed and achieved an accuracy of 98.6% for normal seeds, 98.6% for seeds with fine fissures, and 99.2% for seeds with unclosed glumes. The results showed that the three algorithms achieved desired accuracy.

Analysis of Skin Color Development in Live Goldfish Using a Color Machine Vision System

Abstract A color machine vision system (CMVS) was used to measure and quantify color development in the skin of live goldfish Carassius auratus raised in pond or well water and fed rations of different commercial feeds. Goldfish were of the ornamental Red Oranda variety, 3 months of age, 0.15 ± 0.01 g (mean ± SD) in body weight and 1.38 ± 0.22 cm in total length. Data on color development were acquired through a charge-coupled display color video camera and an image capture card attached to a personal laptop computer. We used a color analysis program developed at the University of Florida, Department of Food Science and Human Nutrition, to interpret the skin color patterns. Pixels from the computerized image were compared and matched to one of 64 color blocks (standards) representing the entire color spectrum. Pixel groupings were represented and recorded as a frequency (percentage) of occurrence for each color standard. The CMVS provided an objective method for measuring the skin color of goldfish and al...

Quality Evaluation of Raw and Cooked Catfish (Ictalurus punctatus) Using Electronic Nose and Machine Vision

Current inspection of catfish quality relies upon sensory evaluation that can be subjective, prone to error and difficult to quantify. Fresh farm-raised catfish (Ictalurus punctatus) fillets were stored at 1.7_ and 7.2_C for up to 6 days and evaluated daily for changes in odors in the raw and cooked stage by an electronic nose as well as by trained sensory panelists. Aerobic microbial counts were performed. Color changes were measured by a color machine vision system. Electronic nose detected odor changes during storage at 1.7_ and 7.2_C. Color analysis showed changes during storage. Addition of color data to electronic nose readings improved quality discrimination among samples. Discriminant function analysis successfully correlated electronic nose sensor and color data with microbial levels, storage time, and sensory readings.

WEED DETECTION USING COLOR MACHINE VISION

Many weed species have reddish stems, but stems of wheat and soybean are green. These color features were used in this study to establish a simple weed-detection method using a color machine-vision system. This method is more practical than texture- or shape-based methods because of its low sensitivity to canopy overlap, leaf orientation, camera focusing, and wind effect. Four types of relative color indices formed by RGB gray levels were designed. The most effective combinations of these color indices were selected using a statistical method. These combinations were used as the input variables for a statistical classifier based on discriminant analysis (DA) and two artificial neural-network (NN) classifiers. These classifiers were trained and tested using three weed species (Johnsongrass, redroot pigweed, and yellow foxtail) with soybean and three weed species (wild buckwheat, cheat, and field bindweed) with wheat. Preprocessing and postprocessing algorithms were developed to shorten the processing time and to reduce noise. The results showed that the statistical DA classifier was more accurate than the NN classifiers in classification accuracy. The least-square means of the classification rates using the DA classifiers for soybean and wheat were 54.9% and 62.2%, respectively. The misclassification rates for most weed species were below 3%. Because the reddish colors on the stems of some weed species vary as the plants grow, an in-field calibration procedure will be needed to make the classifiers more adaptive to different circumstances.

Identification of Damaged Kernels in Wheat using a Colour Machine Vision System

A colour machine vision system was used for identification of healthy and six types of damaged kernels [broken, grass-green/green-frosted, black-point/smudged (selected from grain samples), mildewed, heated and bin-/fire-burnt (created in the laboratory)] of Canada Western Red Spring (CWRS) wheat. A software package was developed to extract various morphological and colour features from the images of both healthy and damaged kernels. Different feature models, morphological, colour and combined (morphological and colour), were evaluated for the identification analysis using the SAS procedures, STEPDISC and DISCRIM. Both parametric and non-parametric statistical classification methods were evaluated with the selected feature models. Colour features proved to be efficient for the identification of healthy and damaged kernels, while combining morphological with colour features improved the identification accuracy. Using a non-parametric classifier with a selected combined model of 24 colour and 4 morphological features, the average identification accuracies were: 93% (healthy), 90% (broken), 99% (grass-green/green-frosted), 99% (black-point/smudged), 99% (mildewed), 98% (heated), and 100% (bin-/fire-burnt), when trained and tested with three different training and testing data sets.

Modelling and evaluating colour information for robot vision

Colour machine vision system characteristics, such as colour balance, illumination, random noise, geometrical distortion and non-linearity in sensitivity, play a crucial role in forming high quality images. High quality input data help alleviate time consuming and computational expensive algorithms. The evaluation and the trade-offs of the aforementioned parameters are asserted in this paper. Colour space issues are addressed and the colour fraction transform is studied over a wide range of lighting conditions. A fast and effective algorithm for colour image segmentation based on colour fractions co-ordinates, including experimental results, is also asserted in this paper.

The Quality of 'Golden Delicious' Apples by Colour Computer Vision

Abstract A colour machine vision system was developed to form a basis for colour grading and defect inspection of 'Golden Delicious' apples. The criteria were based on European Union standards and took into account commercial practices which add subclasses to the basic categories. The system was able to grade correctly more than 90% of the apples for colour (94% by using three colorimetric parameters R,G,B or H,S,I and 91% by using the single canonical variate) and ensured good defect detection (russet, scab, fungi attack, etc.).