Balance between Saturation and Value in Zanthoxylum Bungeanum Fruit Detection

We propose an adaptive threshold method based on the HSV color space to detect the zanthoxylum bungeanum fruit. The thresholds are obtained by keeping the sum of the Saturation and Value components unchangeable. There are many hundreds of pictures for the test dataset. Our method works very well, and the recall rate, accuracy, and false alarm are separately 100%, 100%, and 0%, improved by 5%, 9%, and 58% compared with the fixed thresholds. And to evaluate the location precision in the application of an automatic picking system, we address two metrics: the ratio of the overlap area between the detection region and the ideal region, and the location error. The ratio of the overlap area of our method is 54%, and the location error is about 11%.