Toward Robotic Weed Control: Detection of Nutsedge Weed in Bermudagrass Turf Using Inaccurate and Insufficient Training Data

نویسندگان

چکیده

To enable robotic weed control, we develop algorithms to detect nutsedge from bermudagrass turf. Due the similarity between and background turf, manual data labeling is expensive error-prone. Consequently, directly applying deep learning methods for object detection cannot generate satisfactory results. Building on an instance approach (i.e. Mask R-CNN), combine synthetic with raw train network. We propose algorithm high fidelity data, adopting different levels of annotations reduce cost. Moreover, construct a skeleton-based probabilistic map (NSPM) as neural network input reliance pixel-wise precise labeling. also modify loss function cross entropy Kullback-Leibler divergence which accommodates uncertainty in process. implement proposed compare it both Faster R-CNN R-CNN. The results show that our design can effectively overcome impact imprecise insufficient training sample issues significantly outperform counterpart false negative rate only 0.4%. In particular, reduces time by 95% while achieving better performance if comparing original approach.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

TURF and WEED CONTROL

R field experiments with one of the synthetic plant hormones of growth regulators, conducted by the USGA Green Section in collaboration with the Bureau of Plant Industry of the Dept. of Agriculture, have shown striking promise for the effectiveness of this group of chemicals in the control of clover and some of our more serious turf weeds. The application of aqueous sprays containing a spreader...

متن کامل

utonomous robotic weed control systems : A review

Autonomous robotic weed control systems hold promise toward the automation of one of agriculture’s few remaining unmechanized and drudging tasks, hand weed control. Robotic technology may also provide a means of reducing agriculture’s current dependency on herbicides, improving its sustainability and reducing its environmental impact. This review describes the current status of the four core te...

متن کامل

Evaluation of Imazosulfuron for Yellow Nutsedge (Cyperus esculentus) and Broadleaf Weed Control in Potato

Field studies were conducted in 2007 and 2008 near Nyssa, OR, and Pasco and Paterson, WA to evaluate yellow nutsedge and broadleaf weed control and potato tolerance to imazosulfuron. No injury symptoms from imazosulfuron were evident on potato at Nyssa, whereas in Washington, imazosulfuron caused some chlorosis of potato foliage ranging from 6 to 15% and , 4% at 6 and 15 d after POST applicatio...

متن کامل

Design and Fabrication of a Microwave Weed Killer Device for Weed Control Applications

In this paper, the design and the results of a microwave radiation system for agriculture applications is discussed. The system is fabricated and successfully tested on weed seeds. The device, which uses a commercial 1 kW magentron, proved to be effective for preventing the germination control of popular weeds of Iran. Seven weed species were tested separately by using this system and then the ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: IEEE robotics and automation letters

سال: 2021

ISSN: ['2377-3766']

DOI: https://doi.org/10.1109/lra.2021.3098012