The Use of Image Analysis to Detect Seed Contamination-A Case Study of Triticale

The use of image analysis to study the effect of moisture content on the physical properties of grains

Designing machines and equipment for post-harvest operations of agricultural products requires information about their physical properties. The aim of the work was to evaluate the possibility of introducing a new approach to predict the moisture content in bean and corn seeds based on measuring their dimensions using image analysis using artificial neural networks (ANN). Experimental tests were carried out at three levels of wet basis moisture content of seeds: 9, 13 and 17%. The analysis of the results showed a direct relationship between the wet basis moisture content and the main dimensions of the seeds. Based on the statistical analysis of the seed material, it was shown that the characteristics examined have a normal or close to normal distribution, and the seed material used in the investigation is representative. Furthermore, the use of artificial neural networks to predict the wet basis moisture content of seeds based on changes in their dimensions has an efficiency of 82%. The results obtained from the method used in this work are very promising for predicting the moisture content.

Exploring water-absorbing capacity: a digital image analysis of seeds from 120 wheat varieties

Wheat is a staple food crop that provides a significant portion of the world's daily caloric intake, serving as a vital source of carbohydrates and dietary fiber for billions of people. Seed shape studies of wheat typically involve the use of digital image analysis software to quantify various seed shape parameters such as length, width, area, aspect ratio, roundness, and symmetry. This study presents a comprehensive investigation into the water-absorbing capacity of seeds from 120 distinct wheat lines, leveraging digital image analysis techniques facilitated by SmartGrain software. Water absorption is a pivotal process in the early stages of seed germination, directly influencing plant growth and crop yield. SmartGrain, a powerful image analysis tool, was employed to extract precise quantitative data from digital images of wheat seeds, enabling the assessment of various seed traits in relation to their water-absorbing capacity. The analysis revealed significant transformations in seed characteristics as they absorbed water, including changes in size, weight, shape, and more. Through statistical analysis and correlation assessments, we identified robust relationships between these seed traits, both before and after water treatment. Principal Component Analysis (PCA) and Agglomerative Hierarchical Clustering (AHC) were employed to categorize genotypes with similar trait patterns, providing insights valuable for crop breeding and genetic research. Multiple linear regression analysis further elucidated the influence of specific seed traits, such as weight, width, and distance, on water-absorbing capacity. Our study contributes to a deeper understanding of seed development, imbibition, and the crucial role of water absorption in wheat. These insights have practical implications in agriculture, offering opportunities to optimize breeding programs for improved water absorption in wheat genotypes. The integration of SmartGrain software with advanced statistical methods enhances the reliability and significance of our findings, paving the way for more efficient and resilient wheat crop production. Significant changes in wheat seed shape parameters were observed after imbibition, with notable increases in area, perimeter, length, width, and weight. The length-to-width ratio (LWR) and circularity displayed opposite trends, with higher values before imbibition and lower values after imbibition.

Texture analysis and artificial neural networks for identification of cereals-case study: wheat, barley and rape seeds

The scope of the research comprises an analysis and evaluation of samples of rape, barley and wheat seeds. The experiments were carried out using the author's original research object. The air flow velocities to transport seeds, were set at 15, 20 and 25 m s-1. A database consisting of images was created, which allowed to determine 3 classes of kernels on the basis of 6 research variants, including their transportation way via pipe and the speed of sowing. The process of creating neural models was based on multilayer perceptron networks (MLPN) in Statistica (machine learning). It should be added that the use of MLPN also allowed identification of rape seeds, wheat seeds and barley seeds transported via pipe II at 20 m s-1, for which the lowest RMS was 0.05 and the coefficient of classification accuracy was 0.94.

Application of Machine Learning Using Color and Texture Analysis to Recognize Microwave Vacuum Puffed Pork Snacks

The objective of the study was to create artificial neural networks (ANN) capable of highly efficient recognition of modified and unmodified puffed pork snacks for the purposes of obtaining an optimal final product. The study involved meat snacks produced from unmodified and papain modified raw pork (Psoas major) by means of microwave-vacuum puffing (MVP) under specified conditions. The snacks were then analyzed using various instruments in order to determine their basic chemical composition, color and texture. As a result of the MVP process, the moisture-to-protein ratio (MPR) was reduced to 0.11. A darker color and reduction in hardness of approx. 25% was observed in the enzymatically modified products. Multi-layer perceptron networks (MLPN) were then developed using color and texture descriptor training sets (machine learning), which is undoubtedly an innovative solution in this area.

Autofluorescence-spectral imaging as an innovative method for rapid, non-destructive and reliable assessing of soybean seed quality

In the agricultural industry, advances in optical imaging technologies based on rapid and non-destructive approaches have contributed to increase food production for the growing population. The present study employed autofluorescence-spectral imaging and machine learning algorithms to develop distinct models for classification of soybean seeds differing in physiological quality after artificial aging. Autofluorescence signals from the 365/400 nm excitation-emission combination (that exhibited a perfect correlation with the total phenols in the embryo) were efficiently able to segregate treatments. Furthermore, it was also possible to demonstrate a strong correlation between autofluorescence-spectral data and several quality indicators, such as early germination and seed tolerance to stressful conditions. The machine learning models developed based on artificial neural network, support vector machine or linear discriminant analysis showed high performance (0.99 accuracy) for classifying seeds with different quality levels. Taken together, our study shows that the physiological potential of soybean seeds is reduced accompanied by changes in the concentration and, probably in the structure of autofluorescent compounds. In addition, altering the autofluorescent properties in seeds impact the photosynthesis apparatus in seedlings. From the practical point of view, autofluorescence-based imaging can be used to check modifications in the optical properties of soybean seed tissues and to consistently discriminate high-and low-vigor seeds.

Deep and Machine Learning Using SEM, FTIR, and Texture Analysis to Detect Polysaccharide in Raspberry Powders

In the paper, an attempt was made to use methods of artificial neural networks (ANN) and Fourier transform infrared spectroscopy (FTIR) to identify raspberry powders that are different from each other in terms of the amount and the type of polysaccharide. Spectra in the absorbance function (FTIR) were prepared as well as training sets, taking into account the structure of microparticles acquired from microscopic images with Scanning Electron Microscopy (SEM). In addition to the above, Multi-Layer Perceptron Networks (MLPNs) with a set of texture descriptors (machine learning) and Convolution Neural Network (CNN) with bitmap (deep learning) were devised, which is an innovative attitude to solving this issue. The aim of the paper was to create MLPN and CNN neural models, which are characterized by a high efficiency of classification. It translates into recognizing microparticles (obtaining their homogeneity) of raspberry powders on the basis of the texture of the image pixel.