scholarly journals Accurate Weed Mapping and Prescription Map Generation Based on Fully Convolutional Networks Using UAV Imagery

Sensors ◽  
2018 ◽  
Vol 18 (10) ◽  
pp. 3299 ◽  
Author(s):  
Huasheng Huang ◽  
Jizhong Deng ◽  
Yubin Lan ◽  
Aqing Yang ◽  
Xiaoling Deng ◽  
...  
Keyword(s):  
Weed Management ◽  
Experimental Results ◽  
Convolutional Networks ◽  
Fully Convolutional Networks ◽  
Weed Mapping ◽  
Weed Cover ◽  
Map Generation ◽  
Aerial Vehicle ◽  
Weed Infestation ◽  
And Performance

Chemical control is necessary in order to control weed infestation and to ensure a rice yield. However, excessive use of herbicides has caused serious agronomic and environmental problems. Site specific weed management (SSWM) recommends an appropriate dose of herbicides according to the weed coverage, which may reduce the use of herbicides while enhancing their chemical effects. In the context of SSWM, the weed cover map and prescription map must be generated in order to carry out the accurate spraying. In this paper, high resolution unmanned aerial vehicle (UAV) imagery were captured over a rice field. Different workflows were evaluated to generate the weed cover map for the whole field. Fully convolutional networks (FCN) was applied for a pixel-level classification. Theoretical analysis and practical evaluation were carried out to seek for an architecture improvement and performance boost. A chessboard segmentation process was used to build the grid framework of the prescription map. The experimental results showed that the overall accuracy and mean intersection over union (mean IU) for weed mapping using FCN-4s were 0.9196 and 0.8473, and the total time (including the data collection and data processing) required to generate the weed cover map for the entire field (50 × 60 m) was less than half an hour. Different weed thresholds (0.00–0.25, with an interval of 0.05) were used for the prescription map generation. High accuracies (above 0.94) were observed for all of the threshold values, and the relevant herbicide saving ranged from 58.3% to 70.8%. All of the experimental results demonstrated that the method used in this work has the potential to produce an accurate weed cover map and prescription map in SSWM applications.

scholarly journals Transfer Change Rules from Recurrent Fully Convolutional Networks for Hyperspectral Unmanned Aerial Vehicle Images without Ground Truth Data

2020 ◽  
Vol 12 (7) ◽  
pp. 1099 ◽  
Author(s):  
Ahram Song ◽  
Yongil Kim
Keyword(s):  
Unmanned Aerial Vehicle ◽  
Three Dimensional ◽  
Ground Truth ◽  
Training Data ◽  
Fine Tuning ◽  
Target Domain ◽  
Ground Truth Data ◽  
Convolutional Networks ◽  
Fully Convolutional Networks ◽  
Aerial Vehicle

Change detection (CD) networks based on supervised learning have been used in diverse CD tasks. However, such supervised CD networks require a large amount of data and only use information from current images. In addition, it is time consuming to manually acquire the ground truth data for newly obtained images. Here, we proposed a novel method for CD in case of a lack of training data in an area near by another one with the available ground truth data. The proposed method automatically entails generating training data and fine-tuning the CD network. To detect changes in target images without ground truth data, the difference images were generated using spectral similarity measure, and the training data were selected via fuzzy c-means clustering. Recurrent fully convolutional networks with multiscale three-dimensional filters were used to extract objects of various sizes from unmanned aerial vehicle (UAV) images. The CD network was pre-trained on labeled source domain data; then, the network was fine-tuned on target images using generated training data. Two further CD networks were trained with a combined weighted loss function. The training data in the target domain were iteratively updated using he prediction map of the CD network. Experiments on two hyperspectral UAV datasets confirmed that the proposed method is capable of transferring change rules and improving CD results based on training data extracted in an unsupervised way.

scholarly journals Integration of remote‐weed mapping and an autonomous spraying unmanned aerial vehicle for site‐specific weed management

2019 ◽  
Vol 76 (4) ◽  
pp. 1386-1392 ◽  
Author(s):  
Joseph E Hunter ◽  
Travis W Gannon ◽  
Robert J Richardson ◽  
Fred H Yelverton ◽  
Ramon G Leon
Keyword(s):  
Unmanned Aerial Vehicle ◽  
Weed Management ◽  
Site Specific ◽  
Weed Mapping ◽  
Aerial Vehicle

Spatial Pattern of Weeds Based on Multispecies Infestation Maps Created by Imagery

Weed Science ◽  
2016 ◽  
Vol 64 (3) ◽  
pp. 474-485 ◽  
Author(s):  
Louis Longchamps ◽  
Bernard Panneton ◽  
Robin Reich ◽  
Marie-Josée Simard ◽  
Gilles D. Leroux
Keyword(s):  
Spatial Distribution ◽  
Image Segmentation ◽  
Spatial Pattern ◽  
Weed Management ◽  
Herbicide Application ◽  
Weed Species ◽  
Site Specific ◽  
Weed Cover ◽  
Weed Infestation ◽  
Level Of Aggregation

Weeds are often spatially aggregated in maize fields, and the level of aggregation varies across and within fields. Several annual weed species are present in maize fields before postemergence herbicide application, and herbicides applied will control several species at a time. The goal of this study was to assess the spatial distribution of multispecies weed infestation in maize fields. Ground-based imagery was used to map weed infestations in rain-fed maize fields. Image segmentation was used to extract weed cover information from geocoded images, and an expert-based threshold of 0.102% weed cover was used to generate maps of weed presence/absence. From 19 site-years, 13 (68%) demonstrated a random spatial distribution, whereas six site-years demonstrated an aggregated spatial pattern of either monocotyledons, dicotyledons, or both groups. The results of this study indicated that monocotyledonous and dicotyledonous weed groups were not spatially segregated, but discriminating these weed groups slightly increased the chances of detecting an aggregated pattern. It was concluded that weeds were not always spatially aggregated in maize fields. These findings emphasize the need for techniques allowing the assessment of weed aggregation prior to conducting site-specific weed management.

scholarly journals Automated Counting of Rice Panicle by Applying Deep Learning Model to Images from Unmanned Aerial Vehicle Platform

Sensors ◽  
2019 ◽  
Vol 19 (14) ◽  
pp. 3106 ◽  
Author(s):  
Chengquan Zhou ◽  
Hongbao Ye ◽  
Jun Hu ◽  
Xiaoyan Shi ◽  
Shan Hua ◽  
...  
Keyword(s):  
Deep Learning ◽  
Unmanned Aerial Vehicle ◽  
Field Experiments ◽  
Disease Diagnosis ◽  
Global Scale ◽  
High Definition ◽  
Convolutional Networks ◽  
Fully Convolutional Networks ◽  
Detection And Counting ◽  
Aerial Vehicle

The number of panicles per unit area is a common indicator of rice yield and is of great significance to yield estimation, breeding, and phenotype analysis. Traditional counting methods have various drawbacks, such as long delay times and high subjectivity, and they are easily perturbed by noise. To improve the accuracy of rice detection and counting in the field, we developed and implemented a panicle detection and counting system that is based on improved region-based fully convolutional networks, and we use the system to automate rice-phenotype measurements. The field experiments were conducted in target areas to train and test the system and used a rotor light unmanned aerial vehicle equipped with a high-definition RGB camera to collect images. The trained model achieved a precision of 0.868 on a held-out test set, which demonstrates the feasibility of this approach. The algorithm can deal with the irregular edge of the rice panicle, the significantly different appearance between the different varieties and growing periods, the interference due to color overlapping between panicle and leaves, and the variations in illumination intensity and shading effects in the field. The result is more accurate and efficient recognition of rice-panicles, which facilitates rice breeding. Overall, the approach of training deep learning models on increasingly large and publicly available image datasets presents a clear path toward smartphone-assisted crop disease diagnosis on a global scale.

scholarly journals Vision-based autonomous navigation approach for unmanned aerial vehicle transmission-line inspection

2018 ◽  
Vol 15 (1) ◽  
pp. 172988141775282 ◽  
Author(s):  
Xiaolong Hui ◽  
Jiang Bian ◽  
Xiaoguang Zhao ◽  
Min Tan
Keyword(s):  
Unmanned Aerial Vehicle ◽  
Transmission Lines ◽  
Autonomous Navigation ◽  
Three Dimensional ◽  
Transmission Tower ◽  
Correlation Filters ◽  
Convolutional Networks ◽  
Fully Convolutional Networks ◽  
Perspective Image ◽  
Aerial Vehicle

This article presents an autonomous navigation approach based on a transmission tower for unmanned aerial vehicle (UAV) power line inspection. For this complex vision task, a perspective navigation model, which plays an important role in the description and analysis of the flight strategy, is introduced. Based on the proposed navigation model, valuable cues are excavated from a perspective image, which enhances the capability of the perception of three-dimensional direction and simultaneously improves the safety of intelligent inspection. Specifically, for robust and continuous localization of the transmission tower, a developed detecting-tracking visual strategy—comprised tower detection based on a faster region-based convolutional neural network and tower tracking by kernelized correlation filters—is presented. Further, segmentation by fully convolutional networks is applied to the extraction of transmission lines, from which the vanishing point (VP), an important basis for determining the flight heading, can be obtained. For more robust navigation, the designed scheme addresses the scenario of a nonexistent VP. Finally, the proposed navigation approach and constructed UAV platform were evaluated in a practical environment and achieved satisfactory results. To the best of our knowledge, this article marks the first time that a navigation approach based on a transmission tower is proposed and implemented.

Unmanned Aerial Vehicle for Road Monitoring: Fully Convolutional Networks Approach

2019 ◽  
Vol 12 (6) ◽  
Author(s):  
Laksh kotian ◽  
Asita chheda ◽  
Vaibhav Narwane ◽  
Rakesh Raut
Keyword(s):  
Unmanned Aerial Vehicle ◽  
Convolutional Networks ◽  
Fully Convolutional Networks ◽  
Aerial Vehicle

Platform Independent Analysis of Probabilities on Multithreaded Programs

2013 ◽  
Vol 1 (3) ◽  
pp. 48-65
Author(s):  
Yuting Chen
Keyword(s):  
Operating System ◽  
Static Analysis ◽  
Main Idea ◽  
Experimental Results ◽  
Machine To Machine ◽  
Concurrent Program ◽  
Acceptable Accuracy ◽  
Multithreaded Programs ◽  
Independent Analysis ◽  
And Performance

A concurrent program is intuitively associated with probability: the executions of the program can produce nondeterministic execution program paths due to the interleavings of threads, whereas some paths can always be executed more frequently than the others. An exploration of the probabilities on the execution paths is expected to provide engineers or compilers with support in helping, either at coding phase or at compile time, to optimize some hottest paths. However, it is not easy to take a static analysis of the probabilities on a concurrent program in that the scheduling of threads of a concurrent program usually depends on the operating system and hardware (e.g., processor) on which the program is executed, which may be vary from machine to machine. In this paper the authors propose a platform independent approach, called ProbPP, to analyzing probabilities on the execution paths of the multithreaded programs. The main idea of ProbPP is to calculate the probabilities on the basis of two kinds of probabilities: Primitive Dependent Probabilities (PDPs) representing the control dependent probabilities among the program statements and Thread Execution Probabilities (TEPs) representing the probabilities of threads being scheduled to execute. The authors have also conducted two preliminary experiments to evaluate the effectiveness and performance of ProbPP, and the experimental results show that ProbPP can provide engineers with acceptable accuracy.

scholarly journals Automatic Deep Learning Semantic Segmentation of Ultrasound Thyroid Cineclips using Recurrent Fully Convolutional Networks

IEEE Access ◽  
2020 ◽  
pp. 1-1
Author(s):  
Jeremy M. Webb ◽  
Duane D. Meixner ◽  
Shaheeda A. Adusei ◽  
Eric C. Polley ◽  
Mostafa Fatemi ◽  
...  
Keyword(s):  
Deep Learning ◽  
Semantic Segmentation ◽  
Convolutional Networks ◽  
Fully Convolutional Networks
Sign in / Sign up

Export Citation Format

Share Document