Multispectral imaging is an emerging non-destructive technology. In this work its potential for varietal discrimination and identification of tomato cultivars of Nepal was investigated. Two sample sets were used for the study, one with two parents and their crosses and other with eleven cultivars to study parents and offspring relationship and varietal identification respectively. Normalized canonical discriminant analysis (nCDA) and principal component analysis (PCA) were used to analyze and compare the results for parents and offspring study. Both the results showed clear discrimination of parents and offspring. nCDA was also used for pairwise discrimination of the eleven cultivars, which correctly discriminated upto 100% and only few pairs below 85%. Partial least square discriminant analysis (PLS-DA) was further used to classify all the cultivars. The model displayed an overall classification accuracy of 82%, which was further improved to 96% and 86% with stepwise PLS-DA models on high (seven) and poor (four) sensitivity cultivars, respectively. The stepwise PLS-DA models had satisfactory classification errors for cross-validation and prediction 7% and 7%, respectively. The results obtained provide an opportunity of using multispectral imaging technology as a primary tool in a scientific community for identification/discrimination of plant varieties in regard to genetic purity and plant variety protection/registration.

The purpose of this study was to highlight the use of multispectral imaging in seed quality testing of castor seeds. Visually, 120 seeds were divided into three classes: yellow, grey and black seeds. Thereafter, images at 19 different wavelengths ranging from 375–970 nm were captured of all the seeds. Mean intensity for each single seed was extracted from the images, and a significant difference between the three colour classes was observed, with the best separation in the near-infrared wavelengths. A specified feature (RegionMSI mean) based on normalized canonical discriminant analysis, were employed and viable seeds were distinguished from dead seeds with 92% accuracy. The same model was tested on a validation set of seeds. These seeds were divided into two groups depending on germination ability, 241 were predicted as viable and expected to germinate and 59 were predicted as dead or non-germinated seeds. This validation of the model resulted in 96% correct classification of the seeds. The results illustrate how multispectral imaging technology can be employed for prediction of viable castor seeds, based on seed coat colour.