Estimating canopy cover from color digital camera image of rice field

Canopy cover (CC) is a good predictor variable for plant growth parameters such as leaf area index and aboveground biomass. A nondestructive, low-cost, and convenient method is presented for estimating CC using digital camera image analysis. CC was estimated by the ratio of plant pixels to total pixels of digital camera image of rice field. To determine the criteria for segmenting the rice plant from variable soil background, three mosaic images for rice plant, flooded/bare soil, and algae-infested background were prepared from digital camera images that were taken in various field conditions. An image analysis program was developed in Visual Basic to extract red, green, and blue (RGB) features from the mosaic images, calculate RGB-based color indices, and compute the minimum segmentation error for separating rice plant from background. When judged by the segmentation error, modified excessive green index (MEGI) showed the highest potential for segmenting rice plant from flooded/bare soil background, followed by normalized green (g) and excessive green index (EGI). At the threshold MEGI value of 0.03, the segmentation error was the lowest as 0.13%. Any single index considered was not satisfactory in segmenting rice plant from algae-infested background. However, a discriminant function of 1.2553EGI + 0.01735G − 0.01474B was successful in segmenting rice plant from flooded/bare soil and algaeinfested background with segmentation errors of 0.34 and 1.17%, respectively. CC for four rice varieties from tillering to booting stage was estimated based on the threshold value of MEGI and discriminant function and also manually using commercial software. Both estimates of CC showed good relationship of r2 = 0.94, suggesting that a digital camera could be used efficiently for measuring the CC of rice field.