论文部分内容阅读
介绍了一种用于草莓收获机器人的果实定位和果柄检测方法。利用基于OHTA颜色空间的图像分割方法从背景中分割草莓通过计算二值化草莓blob的惯性主轴来判断草莓的姿态,根据草莓的成熟度实现了果实的选择性采摘。试验证明这种方法的平均判别速度为1 s,果柄误判率为7%,在采摘过程中仅对5%的果实造成损伤,满足草莓机器人的收获速度和精度要求。
A fruit positioning and stalk detection method for strawberry harvesting robot is introduced. Segmentation of Strawberries from Background Based on Image Segmentation Method Based on OHTA Color Space The strawberry’s posture is judged by calculating the inertia principal axis of the binarized strawberry blobs and the selective picking of fruits is achieved according to the maturity of the strawberries. Experiments show that the average discriminating speed of this method is 1 s, the false detection rate of the stalk is 7%, only 5% of the fruits are damaged in the picking process, which meets the requirements of strawberry robots for harvesting speed and precision.