Cereal Grain Classification Research Articles

Automatic classification of non-touching cereal grains in digital images using limited morphological and color features

Classification of cereal grains, namely; barley, oat, rye and wheat (Canada Western Amber Durum (CWAD) and Canada Western Red Spring (CWRS)) was performed using morphological and color features. Grain image boundary contours were extracted from the digital images of kernels, expressed as chain-coded points and then approximated by 13 elliptic Fourier coefficients. After normalization of the rotation and starting point of the contours, symmetrical standard coefficients were determined. The symmetrical Fourier index (SFX) of individual kernels was calculated from the product of the sum of absolute symmetrical coefficients and the circularity (roundness) index. Three geometric features, namely; aspect ratio (AR), major diameter (MD) and roundness (Ceq) were determined using ellipse fitting and Green’s transformation of curve integrals, respectively. The morphological classification model was defined using SFX, AR, MD, and Ceq. The color classification model was defined using color indices of individual kernels, which were calculated from the RGB color values of their images. The classification accuracies of different models were evaluated and compared. The combined model defined by morphological and color features achieved a classification accuracy of 98.5% for barley, 99.97% for CWRS, 99.93% for oat, and 100% for rye and CWAD.

A novel, invariant elliptic Fourier coefficient based classification of cereal grains

An algorithm to classify cereal grains, namely: barley, oats, rye and wheat (Canada Western Amber Durum (CWAD) and Canada Western Red Spring (CWRS)) based on grain kernel shapes using invariant elliptic Fourier descriptors (IEFDs) was developed. Boundary contours were extracted from the digital images of kernels, expressed as chain-coded points and then approximated by 13 Fourier harmonics. After normalisation of the size, rotation and starting point of the contours, the IEFDs were determined. Based on the first three IEFDs, perfect classification was achieved for barley, CWAD, CWRS and rye. The classification accuracies of oats were 99.7% and 100% for the first three and five IEFDs, respectively.

Cereal Grain Classification by Optimal Features and Intelligent Classifiers

The present paper focused on the classification of cereal grains using different classifiers combined to morphological, colour and wavelet features. The grain types used in this study were Hard Wheat, Tender Wheat and Barley. Different types of features (morphological, colour and wavelet) were extracted from colour images using different approaches. They were applied to different classification methods.

IMAGE PROCESSING AND PHYSICO‐MECHANICAL PROPERTIES OF BASIL SEED (OCIMUM BASILICUM)

ABSTRACT Various physical properties of basil seed (Ocimum basilicum) were investigated at a moisture content of 5.32% (wet basis). In order to present a quick and an accurate method for measuring small seed dimensions, the image processing data was compared with the one measured experimentally, and the correlations between these were determined. The average length, width and thickness obtained by image processing and micrometer methods were 3.216, 3.220, 1.838, 1.840, 1.363 and 1.374 mm, respectively. The average projected area, sphericity and roundness evaluated by image processing were 4.407 mm2, 0.95 and 0.54, respectively. However, the average geometric mean diameter, arithmetic mean diameter, surface area and sphericity calculated theoretically based on micrometer data were 2.009, 2.145 mm, 12.796 mm2 and 0.62, respectively. The average unit mass, 1,000 grain mass, volume, true density, bulk density and porosity were measured as 0.0020 and 2.23 g; 7.55 mm3; 1,038; 340.24 kg/m3; and 67.2%, respectively. The friction coefficient ranged from 0.300 for basil seed on glass up to 0.547 on plywood surface. The filling and funning angles of repose were established as 14.64° and 12.35°, respectively. The terminal velocity was 4.26 m/s, and textural properties including rupture force, hardness and toughness were determined as 537.4, 926.13 g and 246.32 g/mm, respectively.PRACTICAL APPLICATIONSAccording to the small size of some seeds, dimension measurement is difficult and time consuming; therefore, developing a rapid, easy, accurate and nondestructive method seems to be necessary. In recent years, computer vision system (CVS) has been increasingly used in the agricultural and food industry. Some examples of CVS application in food industry are: classification of cereal grains, evaluation of pork color, measurement and classification of corn whiteness, automated sorting of pistachio nuts, quality inspection of bakery products, inspection of Golden Delicious apples, detection of pinhole damage in almonds and detection of bones in fish and chicken; therefore, we decided to examine the possibility and accuracy of this technique for measuring seed dimension for the first time.In this study, in addition to introducing the CVS system for dimension measurement, the physico‐mechanical properties of basil seed have been determined, which is necessary for the design of equipment for harvesting, transporting, cleaning, packing, processing, etc.

APPLICATION OF MACHINE VISION FOR DETECTION OF FOREIGN MATTER IN WHEAT GRAINS

With the level of automation at every level of food production and the rates at which food is be-ing produced, it is becoming increasingly more important to have systems that can automatically detect foreign matter along the way. The color scheme could make a computer vision system very practical for foreign object detection and removal. Research at laboratory levels has demonstrated that machine vision is an effective method for clas-sification of cereal grains like wheat. Robust ma-chine vision algorithms have been developed and tested to extract morphological, color and textural features of wheat grains and dockage content. The samples used in this study were bulk images of Egyptian wheat (Sakha8) mixed with known quan-tities of barley, rice and stones (0.5%, 1.0%, 2.0% and 5%). Back propagation neural network (BPNN) and statistical classifiers were used for classifica-tion. Results of the study indicate that classification was reduced from about 97% for wheat mixed with stones to 96% for wheat mixed with rice and 93% for wheat mixed with barley (at 5.0% admix-ture).This trend indicates that the features of 1.0% foreign matter admixture started overlapping with other classes (of admixture). On the other hand, 94%, 95% and 97% of 5% barley, rice and stones admixtures (with wheat) were accurately classified using neural network classifier. Using machine vision system, the detection rate for foreign matter in otherwise clean wheat was 100% with no false positives. This detection scheme was based on a linear feature detector incorporating two orthogonal masks.

Classification of cereal grains using wavelet, morphological, colour, and textural features of non-touching kernel images

Images of non-touching kernels of Canada Western Red Spring (CWRS) wheat, Canada Western Amber Durum (CWAD) wheat, barley, oats, and rye were acquired using an area scan camera. Morphological, colour, textural, and wavelet features were extracted from colour images of cereal grains for classification. A total of 51 morphological features, 93 colour features, 56 textural features, and 135 wavelet features were extracted from each kernel. Linear and quadratic statistical classifiers were used for classification using individual types of features and their combinations to find the best feature set and classification method for improved classification of cereal grains. Combining all morphological, colour, textural and wavelet features gave the best classification using the linear discriminant classifier with a classification accuracy of 99.4% for CWRS wheat, followed by 99.3%, 98.6%, 98.5%, and 89.4% for rye, barley, oats, and CWAD wheat, respectively.

Classification of Bulk Samples of Cereal Grains using Machine Vision

Digital image analysis algorithms were developed to classify bulk samples of Canada Western Red Spring (CWRS) wheat, Canada Western Amber Durum (CWAD) wheat, barley, oats, and rye using textural and colour features. The textural features of bulk samples were extracted from different colours, i.e ., red R , green G , or blue B , and colour band combinations [black & white {(R+G+B)/3}; (3 R+2 G+1 B)/6; (2 R+1 G+3 B)/6; or (1 R+3 G+2 B)/6] of images to determine the colour or colour band combination that gave the highest classification accuracies in cereal grains. The textural features extracted from the red colour band at maximum gray-level value 32 gave the highest classification accuracies in cereal grains (mean accuracy, the average of the classification accuracies of the above-mentioned cereal grains, was 100% when tested on an independent data set that had 10 500 grain kernels). When the original bulk images were partitioned into sub-images and textural or colour features extracted from the sub-images were used, the classification accuracies of cereal grains decreased compared to when the original bulk images were used. The mean accuracy was 100% when colour features of bulk samples were used for classification of cereal grains in an independent data set.

An instrumental system for cereal grain classification using digital image analysis

A digital image processing system is described to facilitate objective inspection and classification of cereal grains. Digitized images of whole grain samples of hard red spring wheat, barley, rye and oats were obtained using a charge-coupled device (CCD) video camera interfaced to a custom-built data-acquisition system. Computer programs were written in assembly language for acquisition and preliminary processing of data from the CCD camera. FORTRAN77 programs were developed for image segmentation and feature extraction. Computed grain features include kernel length, width, area, aspect and thinness ratios, contour length and normalized central moments. Size and shape parameters were evaluated with regard to discrimination ability by stepwise discriminant analysis. Canonical discriminant analysis was applied to visualize cereal class differences and a linear discriminant model was derived. Approximately 1% of over 1100 kernels tested were incorrectly classified among wheat, oats, barley and rye in a four-way admixture. The feasibility of the methodology for instrumental determination of foreign material (‘Besatz’) in grain grading is discussed.