分类： 计算机科学 >> 计算机应用技术 提交时间： 2019-11-23
摘要： Controlling weeds with reduced reliance on herbicides is one of the main challenges to move toward a more sustainable agriculture. Robotic weeding is a thought to be a viable way to reduce the environmental loading of agrochemicals while keeping the operation efficiency high. One of the key technologies for performing robotic weeding is automatic detection of crops and weeds in fields. This paper presents an overview on various methods for detecting plants based on machine vision, mainly concentrating on two main challenges: dealing with changing light and crop/weed discrimination. To overcome the first challenge, both physical and algorithmic methods have been proposed. Physical methods can result in a more cumbersome machine while algorithmic methods are less robust. For crop/weed discrimination, deep-learning-based methods have shown obvious advantages over traditional methods based on hand-crafted features. However, traditional methods still hold some merits that can be leveraged to deep-learning-based methods. With the fast development of hardware technologies, researchers should take full advantage of advanced hardware to ease the algorithm design. In the future, the identification of crops and weeds can be more accurate and fine-grained with the support of online databases and computing resources based on the advances in artificial intelligence and communication technologies.
分类： 机械工程 >> 机械制造自动化 提交时间： 2018-03-16
摘要： A machine-vision-based method of locating crops is described in this research. This method was used to provide real-time positional information of crop plants for a mechanical intra-row weeding robot. Within the normalized red, green, and blue chromatic coordinates (rgb), a modified excess green feature (g-r>T & g-b>T) was used to segment plant material from back ground in color images. The threshold T was automatically selected by the maximum variance (OTSU) algorithm to cope with variable natural light. Taking into account the geometry of the camera arrangement and the crop row spacing, the target regions covering the crop rows were defined based on a pinhole camera model. According to the statistical variation in the pixel histogram in each target region, locations of the crop plants were initially estimated. To obtain the accurate locations of crops, median filtering was conducted locally in the bounding boxes of the crops close to the bottom of the images. For the lateral guidance of the robot, a novel method of calculating lateral offset was proposed based on a simplified match between a template and the detected crops. Field experiments were conducted under three different illumination conditions. The results showed that the accurate identification rates on lettuce, cauliflower and maize were all above 95%. The positional error as within ±15 mm, and the average processing time for a 640×480 image was 31 ms. The method was adequate to meet the technical requirement of the weeding robot, and laid a foundation for robotic weeding in commercial production system.
点击量
待评议
点击量
下载量
评论
