This experiment has been conducted as basic study for development of weed detection technology.Forty-two images obtained from a soybean field were processed.In order to segment the images into the object and background, we proposed improved plant image segmentation based on histogram which requires using a vegetation index and threshold.Through K-means clustering analysis, we built a new vegetation index [(b-a)-0.5(1.4R-G)] 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).The proposed method was applied to weed detection algorithm that uses morphological difference between soybean and weed, such as leaf area and roundness.The accuracy of weed detection is 80%.However, it is hard to apply to the image containing extremely high density weeds.