Application-Specific Evaluation of a Weed-Detection Algorithm for Plant-Specific Spraying

Robotic plant-specific spraying can reduce herbicide usage in agriculture while minimizing labor costs and maximizing yield. Weed detection is a crucial step in automated weeding. Currently, weed detection algorithms are always evaluated at the image level, using conventional image metrics. However, these metrics do not consider the full pipeline connecting image acquisition to the site-specific operation of the spraying nozzles, which is vital for an accurate evaluation of the system. Therefore, we propose a novel application-specific image-evaluation method, which analyses the weed detections on the plant level and in the light of the spraying decision made by the robot. In this paper, a spraying robot is evaluated on three levels: (1) On image-level, using conventional image metrics, (2) on application-level, using our novel application-specific image-evaluation method, and (3) on field level, in which the weed-detection algorithm is implemented on an autonomous spraying robot and tested in the field. On image level, our detection system achieved a recall of 57% and a precision of 84%, which is a lower performance than detection systems reported in literature. However, integrated on an autonomous volunteer-potato sprayer-system we outperformed the state-of-the-art, effectively controlling 96% of the weeds while terminating only 3% of the crops. Using the application-level evaluation, an accurate indication of the field performance of the weed-detection algorithm prior to the field test was given and the type of errors produced by the spraying system was correctly predicted.

[1]  Cesar Pedraza,et al.  A Deep Learning Approach for Weed Detection in Lettuce Crops Using Multispectral Images , 2020, AgriEngineering.

[2]  Sujata Chaudhari,et al.  Yolo Real Time Object Detection , 2020 .

[3]  Huihui Yu,et al.  Applications of Deep Learning for Dense Scenes Analysis in Agriculture: A Review , 2020, Sensors.

[4]  Ross B. Girshick,et al.  Focal Loss for Dense Object Detection , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Joris IJsselmuiden,et al.  Transfer learning for the classification of sugar beet and volunteer potato under field conditions , 2018, Biosystems Engineering.

[6]  Cyrill Stachniss,et al.  Fully Convolutional Networks With Sequential Information for Robust Crop and Weed Detection in Precision Farming , 2018, IEEE Robotics and Automation Letters.

[7]  Rasmus Nyholm Jørgensen,et al.  Weed Growth Stage Estimator Using Deep Convolutional Neural Networks , 2018, Sensors.

[8]  Ali Farhadi,et al.  YOLOv3: An Incremental Improvement , 2018, ArXiv.

[9]  Corné Kempenaar,et al.  Advances in Variable Rate Technology Application in Potato in The Netherlands , 2017, Potato Research.

[10]  Cyrill Stachniss,et al.  Real-Time Semantic Segmentation of Crop and Weed for Precision Agriculture Robots Leveraging Background Knowledge in CNNs , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[11]  J. Hemming,et al.  Field Performance Evaluation Method for a Machine Vision Based Intra Row Hoeing Machine , 2018 .

[12]  Kim Arild Steen,et al.  Estimation of the Botanical Composition of Clover-Grass Leys from RGB Images Using Data Simulation and Fully Convolutional Neural Networks , 2017, Sensors.

[13]  Cyrill Stachniss,et al.  Effective Vision‐based Classification for Separating Sugar Beets and Weeds for Precision Farming , 2017, J. Field Robotics.

[14]  Cyrill Stachniss,et al.  Semi-supervised online visual crop and weed classification in precision farming exploiting plant arrangement , 2017, 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[15]  Cyrill Stachniss,et al.  REAL-TIME BLOB-WISE SUGAR BEETS VS WEEDS CLASSIFICATION FOR MONITORING FIELDS USING CONVOLUTIONAL NEURAL NETWORKS , 2017 .

[16]  Rasmus Nyholm Jørgensen,et al.  RoboWeedSupport - Detection of weed locations in leaf occluded cereal crops using a fully convolutional neural network , 2017 .

[17]  Henrik Skov Midtiby,et al.  Plant species classification using deep convolutional neural network , 2016 .

[18]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[19]  Robert Fitch,et al.  Real-time target detection and steerable spray for vegetable crops , 2015 .

[20]  Pietro Perona,et al.  Microsoft COCO: Common Objects in Context , 2014, ECCV.

[21]  Jörn Ostermann,et al.  Plant classification system for crop /weed discrimination without segmentation , 2014, IEEE Winter Conference on Applications of Computer Vision.

[22]  Gerrit Polder,et al.  A robot to detect and control broad‐leaved dock (Rumex obtusifolius L.) in grassland , 2011, J. Field Robotics.

[23]  J. Hemming,et al.  Image analysis system to determine crop row and plant positions for an intra-row weeding machine , 2011 .

[24]  A. T. Nieuwenhuizen,et al.  Performance evaluation of an automated detection and control system for volunteer potatoes in sugar beet fields , 2010 .

[25]  L. Firbank,et al.  Identifying and managing the conflicts between agriculture and biodiversity conservation in Europe - a review , 2008 .

[26]  W. S. Lee,et al.  Robotic Weed Control System for Tomatoes , 2004, Precision Agriculture.

[27]  Clement A. Tisdell,et al.  Why farmers continue to use pesticides despite environmental, health and sustainability costs , 2001 .

[28]  J. Goudriaan,et al.  Design of weed management systems with a reduced reliance on herbicides poses new challenges and prerequisites for modeling crop–weed interactions , 2000 .

[29]  S. Christensen,et al.  Colour and shape analysis techniques for weed detection in cereal fields , 2000 .

[30]  Rainer H. Biller,et al.  Reduced input of herbicides by use of optoelectronic sensors , 1998 .

[31]  G. van Straten,et al.  Intra-Row Weed Control: A Mechatronics Approach , 1998 .

[32]  Reyer Zwiggelaar,et al.  A review of spectral properties of plants and their potential use for crop/weed discrimination in row-crops , 1998 .

[33]  G. Meyer,et al.  Color indices for weed identification under various soil, residue, and lighting conditions , 1994 .

[34]  Gaines E. Miles,et al.  MACHINE VISION AND IMAGE PROCESSING FOR PLANT IDENTIFICATION. , 1986 .