Sensing tomato’s pathogen using Visible/Near infrared (VIS/NIR) spectroscopy and multivariate data analysis (MVDA)

Abstract

Quality of agricultural products is a very important issue for consumers as well as for farmers in relation to price, health and flavour. One of the factors that determine the quality is the absence of pathogens that can cause diseases for products and also for consumers. An advanced method to sense pathogens and their antagonists is the use of Visible/Near Infrared (VIS/NIR) spectroscopy. In this paper, the VIS/NIR spectroscopy, with the help of two techniques of multivariate data analysis (MVDA); namely principal component analysis (PCA) and support vector machine (SVM)-classification; showed very reliable results for sensing two artificially inoculated fungi (Fusarium oxysporum f. sp. Lycopersici and Rhizoctonia solani), and two antagonistic bacteria (Bacillus atrophaeus and Pseudomonas aeruginosa). The two fungi cause loss of quality and quantity for tomatoes. The results showed that the lowest classification rates using VIS/NIR spectroscopy for pathogens, antagonistic and their combinations were 90%, 85% and 74%, respectively. These results open a wide range for using VIS/NIR spectroscopy sensor technology for agricultural commodities quality at quality control checkpoints.

Recent advancements in NIR spectroscopy for assessing the quality and safety of horticultural products: A comprehensive review

“…During spectral analysis, this method showed a spectral shift due to the leaking from cabbage cells and the growth of bacteria ( 82 ). This shifting behavior was treated using the SG-derivative method ( 82 ), followed by tomato pathogen analysis ( 83 ). Spectral analysis revealed that the shredded samples had better separation between inoculation times than the ground samples ( 82 ), which makes the former the method of choice for analysis.…”

“…A study on tomato pathogen analysis achieved a classification accuracy rate of 99.3% on the first and second day using PCA followed by an SVM classifier ( 83 ). Although higher classification rates were achieved as the mixing of samples began, the SVM classification could not handle multiple factors affecting the same.…”

“…Although higher classification rates were achieved as the mixing of samples began, the SVM classification could not handle multiple factors affecting the same. The high classification rates were able to detect F. oxysporum and R. solani fungi, as well as bacteria such as B. atrophaeus and P. aeruginosa ( 83 ).…”

“…NIR essentially works through photons and their response to corresponding compounds, so as a solution becomes clear, it becomes non-homogeneous and therefore changes the spectral response. In this context, it is best to use the SW-NIR technique for analysis due to its higher penetration power and lesser interference from water bands ( 83 ). However, the study on microbial contamination in lettuce suggested that the water band corresponding to 970 nm O-H stretching needs to be accounted for ( 36 ).…”

Suitability of the MODIS-NDVI Time-Series for a Posteriori Evaluation of the Citrus Tristeza Virus Epidemic

“…For example, research conducted by CIHEAM of Bari showed the application of RS in the monitoring of two quarantine pathogens in the Apulia region (Italy): Citrus tristeza virus (CTV) and Xylella fastidiosa [4]. Visible/near-infrared (VIS/NIR) and thermal (TIR) imaging techniques have been applied for detecting the stress condition of plants infected by pathogens, such as fungi [5,6], bacteria [7] and viruses, e.g., tobacco mosaic virus [8], grape leaf roll-associated virus-3 [9] or sugarcane yellow leaf virus [10].…”