Vegetation segmentation in a field color image is a process of distinguishing vegetation objects of interests like crops and weeds from a background of soil and/or other residues. The performance of the process is crucial in automatic precision agriculture which includes weed control and crop status monitoring. To facilitate the segmentation, color indices have predominantly been used to transform the color image into its gray-scale image. A thresholding technique like the Otsu method is then applied to distinguish vegetation parts from the background. An obvious demerit of the thresholding based segmentation will be that classification of each pixel into vegetation or background is carried out solely by using the color feature of the pixel itself without taking into account color features of its neighboring pixels. This paper presents a new pixel-based segmentation method which employs a multi-layer perceptron neural network to classify the gray-scale image into vegetation and nonvegetation pixels. The input data of the neural network for each pixel are 2-dimensional gray-level values surrounding the pixel. To generate a gray-scale image from a raw RGB color image, a well-known color index called Excess Green minus Excess Red Index was used. Experimental results using 80 field images of 4 vegetation species demonstrate the superiority of the neural network to existing threshold-based segmentation methods in terms of accuracy, precision, recall, and harmonic mean.

We address improved plant image segmentation based on histograms which requires using a vegetation index and threshold. Image segmentation is the most important step for extracting targets, such as vegetation, from images; this affects successful detection of plant information. Forty-two field images were acquired from a soybean field using an RGB camera. Through K-means clustering analysis, we built a new vegetation index        and generated gray-scale images. Otsu and Triangle thresholds were used to convert contrast images to binary. Optimal threshold values were generally located between the Otsu and Triangle threshold values. The combined threshold method shows 98.79% and 0.95% of mean accuracy and standard deviation, respectively, whereas the Otsu and Triangle method results show 98.17±1.71% and 97.85±1.87%, respectively. These results show that the combined method has significant segmentation potential through one-way ANOVA. Then we compared the results with K-means clustering using two-sample t-test. The K-means method’s mean accuracy is 98.18±1.79%, with no significant difference between the proposed and K-means methods. However, the proposed method’s processing time is 0.60±0.01 s, i.e., twice faster than the K-means method (1.72±0.24 s).