Discrimination of Kernel Quality Characteristics for Sunflower Seeds Based on Multispectral Imaging Approach

Abstract

Multispectral imaging in the visible and nearinfrared (405-970 nm) regions was tested for nondestructive discrimination of insect-infested, moldy, heterochromatic, and rancidity in sunflower seeds. An excellent classification (accuracy >97 %) for intact sunflower seeds could be achieved using Fisher's linear discriminant function based on 10 feature wavelengths that were selected from the original 19 wavelengths by Wilks' lambda stepwise method. Intact sunflower seeds with different degree of rancidity could be precisely clustered by multispectral imaging technology combined with principal component analysis-cluster analysis (PCA-CA). Our results demonstrate the capability of multispectral imaging technology as a tool for rapid and nondestructive analysis of seed quality attributes, which enables many applications in the agriculture and food industry.

Multispectral imaging – a new tool in seed quality assessment?

“…A further application of MSI is the determination of insect damage (Ma et al , 2015; Sendin et al ., 2018) and other crop and weed seeds (Sendin et al ., 2018) have recently been documented.…”

Feasibility of combining spectra with texture data of multispectral imaging to predict heme and non-heme iron contents in pork sausages

“…The spatial position of each pixel along with its iron content was then used to form the distribution map of iron within PS. The detailed steps involved in visualization were depicted in previous study (Ma et al, 2014a). …”

“…Compared with hyperspectral imaging, less processing time make MSI easer to meet the speed requirement of industrial production lines used for online high-throughput screening to assess meat quality, nutrition, safety, and authenticity. In recent years, MSI technology as a rapid and nondestructive analysis method has been wildly applied in fishes (Dissing, Nielsen, Ersbøll, & Frosch, 2011), meats (Ma et al, 2014a;Ropodi, Pavlidis, Mohareb, Panagou, & Nychas, 2015), fruits (Lunadei, Galleguillos, Diezma, Lleó, & Ruiz-Garcia, 2011), vegetables (Løkke, Seefeldt, Skov, & Edelenbos, 2013), etc. However, almost all of these studies only utilized spectral data without incorporating information on spatial data, which has been successfully used to improve the accuracy of model prediction in hyperspectral imaging studies (He, Wu, & Sun, 2014;Pu, Sun, Ma, & Cheng, 2015).…”

Multispectral Imaging for Plant Food Quality Analysis and Visualization

“…The highest classification accuracy of 82% was obtained for insect‐infested apples (Rady and others ). The multispectral imaging technique has also been widely investigated to detect various types of defects (such as insect damage, bruising, decay, cold injury, black heart, puncture injury, and cracks) on various plant foods (such as peach, radish, sunflower seed, citrus, and jujube) (Ma and others ; Zhang and others ; Folch‐Fortuny and others ; Li and others , ; Liu and others ; Pan and others ; Song and others ; Wu and others ). Based on feature wavelengths associated with corresponding defects, simplified models (such as soft independent modeling of class analogy (SIMCA), PCA, ANN, LS‐SVM, FLDA, and MNF) were conducted for nondestructively assessing defects on such plant foods with classification accuracies of over 90%.…”