UAV-based crop and weed classification for smart farming

Lottes, Philipp; Khanna, Raghav; Pfeifer, Johannes; Siegwart, Roland; Stachniss, Cyrill

doi:10.1109/icra.2017.7989347

Cited by 299 publications

(177 citation statements)

References 16 publications

Supporting

Mentioning

172

Contrasting

Unclassified

Order By: Relevance

“…In the field of machine learning, several approaches have been applied to classify crops and weeds in imagery of plantation (GarciaRuiz et al, 2015;Guerrero et al, 2012;Perez-Ortiz et al, 2016Haug et al, 2014;Lottes et al, 2016aLottes et al, , 2017Latte et al, 2015). Perez-Ortiz et al (2015) propose an image patch based weed detection system.…”

Section: Related Workmentioning

confidence: 99%

“…They perform a prediction on pixel level on a sparse grid in image space and report an average accuracy of around 94% on an evaluation set of 70 images where both, intra and interrow overlap is present. In previous work, Lottes et al (2016aLottes et al ( , 2017 design a crop and weed classification system for ground and aerial robots. The work exploits NDVI and ExG indices to first segment the vegetation in the image data and then applied a random forest classifier to the vegetative parts to further distinguish them into crops and weeds.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Real-Time Blob-Wise Sugar Beets vs Weeds Classification for Monitoring Fields Using Convolutional Neural Networks

Milioto

Lottes

Stachniss

2017

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

116

View full text Add to dashboard Cite

ABSTRACT:UAVs are becoming an important tool for field monitoring and precision farming. A prerequisite for observing and analyzing fields is the ability to identify crops and weeds from image data. In this paper, we address the problem of detecting the sugar beet plants and weeds in the field based solely on image data. We propose a system that combines vegetation detection and deep learning to obtain a high-quality classification of the vegetation in the field into value crops and weeds. We implemented and thoroughly evaluated our system on image data collected from different sugar beet fields and illustrate that our approach allows for accurately identifying the weeds on the field.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Real-Time Blob-Wise Sugar Beets vs Weeds Classification for Monitoring Fields Using Convolutional Neural Networks

Milioto

Lottes

Stachniss

2017

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

116

View full text Add to dashboard Cite

show abstract

“…A smart farming system using unmanned aerial vehicles is proposed by P. Lottes et al [8]. The authors described the proposed system for vegetation detection, plant-tailored feature extraction, and classification to obtain an estimate of the distribution of crops and weed relying on objectfeatures and key points in combination with random forest and showed the classification system achieve good performance.…”

Section: Related Workmentioning

confidence: 99%

Automated Aeroponics System Using IoT for Smart Farming

Kerns¹,

Lee²

2017

ESJ

View full text Add to dashboard Cite

In recent years, the Internet of Things (IoT) has received much attention in the areas of industry and academia. Currently, IoT technologies are being applied in many fields and is changing lives in many areas such as smart homes, smart cities, smart grids, autonomous cars, and the industrial internet. Howbeit, traditional agriculture is still waiting for many changes to occur in networking technology especially in IoT. Many researchers and engineers are working towards applying IoT technology to traditional agricultural methods. Aeroponics farming is an efficient and effective process for growing plants without using soil. When we apply IoT technology to an aeroponics system, it is expected that there will be many improvements such as decreasing water usage, increasing plant yield, minimizing rate of growth and reducing the workforce. In this paper, we designed and implemented a new automatic aeroponics system using IoT devices. Our system is comprised of three main components: a mobile application, service platform and IoT devices with sensors. The mobile application provides the user a graphical user interface to monitor and adjust the aeroponics system. The service platform is a middleware system that provides information for the mobile application to store the gathered information from IoT devices using sensors within the aeroponics system. The IoT device uses sensors within the aeroponics system to control each pump and access data. Our work is a new application in the agricultural industry and is expected to be a promising application that will help the farmer with increasing productivity in farming and reducing carbon footprint. IntroductionAs technology develops, it will bring great advantages to human beings in all areas of life. Especially, the development of network technology is changing many areas of our lives such as smart homes, smart cities, smart grids, autonomous cars, and the industrial internet. Smart agriculture systems are an area that would benefit from this technology. By adapting the recent innovations in technology for farming it would add many benefits for farmers and reduce our carbon footprint.The Internet of Things (IoT) [1] is an emerging technology that provides internet connectivity and powerful data analytic capabilities to create a system in which objects in the physical world could be connected to the internet by sensors. However, smart farming is still in its infancy in regard to using IoT devices for agriculture. Aeroponics farming is an efficient and effective process for growing plants without using soil as a medium as well as with little water. In this paper, we propose new automated aeroponics system using IoT devices that will help the farmer by increasing productivity in farming. To the best of our knowledge, this is first work to design and implement of automated aeroponics system using IoT devices.The rest of the paper is organized as follows. Section II describes current research work in smart farming systems with IoT devices and IoT applications. In secti...

show abstract

“…Several studies have demonstrated that the choice of at least four spectral bands provided better results than the use of standard RGB camera especially when near infra-red (NIR) information was used [20,21]. In particular, the use of NIR information improves the separation between the vegetation and the background (i.e., soil) in image pre-processing [22]. In recent studies, Pérez-Ortiz, et al [23] used an RGB camera, Peña, et al [24] used a six-band multispectral camera (i.e., bandpass filters centred at 530, 550, 570, 670, 700, and 800 nm), López-Granados, et al [25] used an RGB camera and a six-band multispectral camera (i.e., bandpass filters centred at 450, 530, 670, 700, 740, and 780 nm).…”

Section: Introductionmentioning

confidence: 99%

“…A first automatic labelling step was performed to build the training data; nevertheless a manual verification of the labelled patterns is still required for these data. Lottes, Khanna, Pfeifer, Siegwart and Stachniss [22] used an object-based and a keypoint-based approach. They proposed a classification system taking into account several spectral and geometrical features.…”

Section: Introductionmentioning

confidence: 99%

Unsupervised Classification Algorithm for Early Weed Detection in Row-Crops by Combining Spatial and Spectral Information

Louargant

Jones

Faroux³

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

Abstract:In agriculture, reducing herbicide use is a challenge to reduce health and environmental risks while maintaining production yield and quality. Site-specific weed management is a promising way to reach this objective but requires efficient weed detection methods. In this paper, an automatic image processing has been developed to discriminate between crop and weed pixels combining spatial and spectral information extracted from four-band multispectral images. Image data was captured at 3 m above ground, with a camera (multiSPEC 4C, AIRINOV, Paris) mounted on a pole kept manually. For each image, the field of view was approximately 4 m × 3 m and the resolution was 6 mm/pix. The row crop arrangement was first used to discriminate between some crop and weed pixels depending on their location inside or outside of crop rows. Then, these pixels were used to automatically build the training dataset concerning the multispectral features of crop and weed pixel classes. For each image, a specific training dataset was used by a supervised classifier (Support Vector Machine) to classify pixels that cannot be correctly discriminated using only the initial spatial approach. Finally, inter-row pixels were classified as weed and in-row pixels were classified as crop or weed depending on their spectral characteristics. The method was assessed on 14 images captured on maize and sugar beet fields. The contribution of the spatial, spectral and combined information was studied with respect to the classification quality. Our results show the better ability of the spatial and spectral combination algorithm to detect weeds between and within crop rows. They demonstrate the improvement of the weed detection rate and the improvement of its robustness. On all images, the mean value of the weed detection rate was 89% for spatial and spectral combination method, 79% for spatial method, and 75% for spectral method. Moreover, our work shows that the plant in-line sowing can be used to design an automatic image processing and classification algorithm to detect weed without requiring any manual data selection and labelling. Since the method required crop row identification, the method is suitable for wide-row crops and high spatial resolution images (at least 6 mm/pix).

show abstract

UAV-based crop and weed classification for smart farming

Cited by 299 publications

References 16 publications

Real-Time Blob-Wise Sugar Beets vs Weeds Classification for Monitoring Fields Using Convolutional Neural Networks

Real-Time Blob-Wise Sugar Beets vs Weeds Classification for Monitoring Fields Using Convolutional Neural Networks

Automated Aeroponics System Using IoT for Smart Farming

Unsupervised Classification Algorithm for Early Weed Detection in Row-Crops by Combining Spatial and Spectral Information

Contact Info

Product

Resources

About