Plant wilt detection by computer-vision tracking of leaf tips

The vertical movement of leaf-tips of four plants were tracked simultaneously by a computer vision system. Fully expanded leaves of tomato plants were found to have linear vertical motions in response to both water stress level and carbon dioxide assimilation rate. Growing leaves had complex motions which were less useful for monitoring water stress level. The computer vision detected the onset of wilt before physiological injury took place and triggered irrigation at predetermined leaf tip deflections.