In order to eliminate the impact of the illumination on the tomato image segmentation, the paper adopts a new method. The method is based on illumination irrelevant image, and uses minimum entropy criterion to calculate the illumination irrelevant angle of the given camera. We preprocess the colorful tomato image by the efficient median filter method, and obtain the illumination irrelevant image according to the formative principle of images. Then we segment the illumination irrelevant images using the improved Ostu segmentation method, and compare with the result of the segmentation images based on the chromatic aberration method. We also adopt the statistical threshold method in HIS color space to segment tomato images and study the influence of illumination. The experiment shows that the illumination irrelevant method can eliminate the influence of illumination effectively, and segments the objects accurately with the distinguished features.

In this study, Ethylene-vinyl acetate (EVA) based composites filament, with four different volume fraction of the nano-sized carbon black particles (NCB) were produced by melt mixing using a single extruder. The morphology of the EVA/NCB was studied using SEM, where a 3-D network of the NCB was presented. Thermal gravity analysis (TGA) measurement was utilized, denoting the degradation temperature of EVA, and presetting the actual volume fraction of NCB in the composites. Mechanical properties, e.g. elongation at break, tensile strength of the EVA/NCB filament was studied. Most importantly, the strain sensoring behavior of the EVA/NCB was investigated ultilizing a tensile testing machine coupled with a pico-ammeter. The gauge factor for various strain range, as well as the relative change of the resistance during the cyclic measurement of the NCB/EVA composites was calculated. Moreover, the measured data were fitted using some mathematical modellings, which reveals the potential of the strain sensoring behavior of the NCB/EVA composites in this study. Overall, this study introduces a durable NCB/EVA composites using as strain sensor oriented to industrial large-scale production, and the proposed modelling provides an effective evaluation method on its strain sensoring behavior.

No abstract

In order to improve the recognition accuracy of vision system on tomato picking robots, the paper proposed a method of feature extraction and recognition for ripe tomato based on illumination irrelevant images and support vector machine (SVM). In this method, we adopted vector median filter (VMF)