نوع مقاله : مقاله پژوهشی
نویسندگان
1 گروه مهندسی بیوسیستم، دانشکده کشاورزی، دانشگاه بوعلی سینا، همدان، ایران.
2 استادیار، گروه مهندسی بیوسیستم، دانشکده کشاورزی، دانشگاه بوعلی سینا، همدان، ایران
3 دانشیار، گروه مهندسی بیوسیستم، دانشکده کشاورزی، دانشگاه ایلام، ایلام، ایران
چکیده
کلیدواژه‌ها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Quality assessment of powdery products requires laboratory methods that are complicated, expensive, and time-consuming. Therefore, the purpose of the present research was to implement a simple, cheap, and fast system for the quality of powder products. In the present research, hyperspectral imaging and image processing technologies were used. The goal product was red pepper powder. Wheat flour, chickpea flour, and sea foam powder with different mass percentages (0-50%) were mixed with red pepper as adulteration. A hyperspectral imaging camera and image processing technology were used. In order to process images, an algorithm was implemented in MATLAB software. Principal component analysis method was used to determine the effective wavelengths. After extracting the features of the images, the sequential feature selection method was used to determine the efficient features and to be classified using the support vector machine method. The effective wavelengths for chickpea flour were 530.22, 633.57, 704.68, 793.98, 844.93, and 899.81 nm, for sea foam were 509.55, 607.11, 626.96, 769.17, 813, and 867.57 nm and for wheat flour adulteration were 518.4, 607.11, 705.51, 794.81, 855.17, and 909.74 nm. The number of efficient features to detect adulteration of chickpea flour, sea foam powder, and wheat flour in red pepper powder were 12, 12, and 10, respectively. The support vector machine with the one-against-one method was more efficient than the one-against-all method and its correct classification rates in detecting the adulterations were 97.77, 95.55, and 95.55%, respectively.
کلیدواژه‌ها [English]
)