Puccinia recondita f. sp. tritici, the causative agent of brown rust disease is the reason for major yield losses in winter wheat plants (Triticum aestivum L.). Early detection of fungal infections would allow a specific fungicide application. Aim of the current study was to develop an automated and computer based device that differentiate infected and healthy plants non-invasively in an early stage of infection. To achieve the target an optical sensor device has been designed (“MultiDetExc”), which excites chlorophyll fluorescence in discrete wavelengths and detects the induced emissions. Wheat plants were infected artificially with brown rust in a climate chamber experiment to survey the capability of the instrument. The chlorophyll fluorescence excitation spectra of whole wheat plants were measured on several days after infection. As reference methods, HPLC and qPCR analysis were included in the current study to measure the polyphenol content of the leaves and the level of infection.The recently developed sensor device is an efficient technique to differentiate the infected and not infected wheat plants as soon as four days after inoculation. The measured fluorescence quotients correlate high positive with the polyphenol contents and the relative amount of fungal DNA. An untreated healthy control was measured parallelly, in order to associate the increased synthesis of polyphenols to the fungal infection definitely.