Development of a visible-NIR sensor for potato late blight detection

点击翻译按钮获取中文摘要

A&L Canada Labs Inc. is developing an UAV (Unmanned Aerial Vehicle) with a multi-spectral camera to determine nutrient (nitrogen, phosphorus, potassium, magnesium) levels, detect diseases, and to assess overall health of specific crops. The multi-spectral camera will have 6 radiation absorption bands: one in the red band, one in the blue band, one in the green band, and three in the NIR regions. The purpose of the study is to determine which near-infrared (NIR) bands are the most suitable to detect late blight disease in potato fields located in Prince Edward Island. Late blight is due to Phytophthora infestans, and devastates potato foliage and tuber. Late blight weather-based forecasting systems have been developed as it most commonly occurs in cool, wet climates, but the systems give general predictions and a more precise crop assessment is needed. One way to achieve such an assessment is through aerial detection of the diseases using non-destructive radiation based measurements. In the visible (400-700 nm) and NIR regions (700-1900 nm) of the electromagnetic spectrum, there are many spectral signatures that can be related to late blight diseases in potato or in tomato. The challenge is to identify which spectral bands are the best to detect P. infestans in the case of potato cultivars grown in Prince Edward Island. For each band, band width and locations will have to be determined.