Applications of Deep Learning for Dense Scenes Analysis in Agriculture: A Review

Zhang, Qian; Liu, Yeqi; Gong, Chuanyang; Chen, Yingyi; Yu, Huihui

doi:10.3390/s20051520

Cited by 134 publications

(98 citation statements)

References 116 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As for the image itself, the major limitation of a UAV image data collection is the low capacity to compensate and analyze larger areas. However, this type of aerial remote sensing is important when considering the spatial resolution and highly detailed information obtained on the vegetation cover, permitting an analysis at a plant or crop-plot level [68][69][70][71]. Additionally, by evaluating crop at an aerial view, it is easier to ascertain the relationship between spectral data and biophysical variables, since the end-user can reduce the amount of noise introduced in the system by extracting only pixels corresponding with the canopy itself.…”

Section: Discussionmentioning

confidence: 99%

Leaf Nitrogen Concentration and Plant Height Prediction for Maize Using UAV-Based Multispectral Imagery and Machine Learning Techniques

et al. 2020

View full text Add to dashboard Cite

Under ideal conditions of nitrogen (N), maize (Zea mays L.) can grow to its full potential, reaching maximum plant height (PH). As a rapid and nondestructive approach, the analysis of unmanned aerial vehicles (UAV)-based imagery may be of assistance to estimate N and height. The main objective of this study is to present an approach to predict leaf nitrogen concentration (LNC, g kg−1) and PH (m) with machine learning techniques and UAV-based multispectral imagery in maize plants. An experiment with 11 maize cultivars under two rates of N fertilization was carried during the 2017/2018 and 2018/2019 crop seasons. The spectral vegetation indices (VI) normalized difference vegetation index (NDVI), normalized difference red-edge index (NDRE), green normalized difference vegetation (GNDVI), and the soil adjusted vegetation index (SAVI) were extracted from the images and, in a computational system, used alongside the spectral bands as input parameters for different machine learning models. A randomized 10-fold cross-validation strategy, with a total of 100 replicates, was used to evaluate the performance of 9 supervised machine learning (ML) models using the Pearson’s correlation coefficient (r), mean absolute error (MAE), coefficient of regression (R²), and root mean square error (RMSE) metrics. The results indicated that the random forest (RF) algorithm performed better, with r and RMSE, respectively, of 0.91 and 1.9 g.kg−¹ for LNC, and 0.86 and 0.17 m for PH. It was also demonstrated that VIs contributed more to the algorithm’s performances than individual spectral bands. This study concludes that the RF model is appropriate to predict both agronomic variables in maize and may help farmers to monitor their plants based upon their LNC and PH diagnosis and use this knowledge to improve their production rates in the subsequent seasons.

show abstract

Section: Discussionmentioning

confidence: 99%

Leaf Nitrogen Concentration and Plant Height Prediction for Maize Using UAV-Based Multispectral Imagery and Machine Learning Techniques

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Deep learning object detection algorithms outperform handcrafted algorithms in the agricultural domain [ 19 , 21 , 24 ]. Therefore, in this research, a deep learning algorithm was used for the weed detection task.…”

Section: Methodsmentioning

confidence: 99%

“…In more recent studies, performance could be further improved by using deep-learning algorithms. These algorithms learn the relevant image features to detect weeds directly from the camera images [ 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 ].…”

Section: Introductionmentioning

confidence: 99%

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

Ruigrok

Henten

Booij

et al. 2020

Sensors

View full text Add to dashboard Cite

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.

show abstract

“…Instance segmentation has applications in several areas of knowledge: medicine [69,70], biology [71,72], livestock [73,74], agronomy [75,76], among others. However, remote sensing application is still restricted, highlighting its use in the automatic detection of the following targets: marine oil spill [77], building [78,79], vehicle [80], and ship [81].…”

Section: Introductionmentioning

confidence: 99%

Instance Segmentation for Large, Multi-Channel Remote Sensing Imagery Using Mask-RCNN and a Mosaicking Approach

et al. 2020

View full text Add to dashboard Cite

Instance segmentation is the state-of-the-art in object detection, and there are numerous applications in remote sensing data where these algorithms can produce significant results. Nevertheless, one of the main problems is that most algorithms use Red, Green, and Blue (RGB) images, whereas Satellite images often present more channels that can be crucial to improve performance. Therefore, the present work brings three contributions: (a) conversion system from ground truth polygon data into the Creating Common Object in Context (COCO) annotation format; (b) Detectron2 software source code adaptation and application on multi-channel imagery; and (c) large scene image mosaicking. We applied the procedure in a Center Pivot Irrigation System (CPIS) dataset with ground truth produced by the Brazilian National Water Agency (ANA) and Landsat-8 Operational Land Imager (OLI) imagery (7 channels with 30-m resolution). Center pivots are a modern irrigation system technique with massive growth potential in Brazil and other world areas. The round shapes with different textures, colors, and spectral behaviors make it appropriate to use Deep Learning instance segmentation. We trained the model using 512 × 512-pixel sized patches using seven different backbone structures (ResNet50- Feature Pyramid Network (FPN), Resnet50-DC5, ResNet50-C4, Resnet101-FPN, Resnet101-DC5, ResNet101-FPN, and ResNeXt101-FPN). The model evaluation used standard COCO metrics (Average Precision (AP), AP50, AP75, APsmall, APmedium, and AR100). ResNeXt101-FPN had the best results, with a 3% advantage over the second-best model (ResNet101-FPN). We also compared the ResNeXt101-FPN model in the seven-channel and RGB imagery, where the multi-channel model had a 3% advantage, demonstrating great improvement using a larger number of channels. This research is also the first with a mosaicking algorithm using instance segmentation models, where we tested in a 1536 × 1536-pixel image using a non-max suppression sorted by area method. The proposed methodology is innovative and suitable for many other remote sensing problems and medical imagery that often present more channels.

show abstract

Applications of Deep Learning for Dense Scenes Analysis in Agriculture: A Review

Cited by 134 publications

References 116 publications

Leaf Nitrogen Concentration and Plant Height Prediction for Maize Using UAV-Based Multispectral Imagery and Machine Learning Techniques

Leaf Nitrogen Concentration and Plant Height Prediction for Maize Using UAV-Based Multispectral Imagery and Machine Learning Techniques

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

Instance Segmentation for Large, Multi-Channel Remote Sensing Imagery Using Mask-RCNN and a Mosaicking Approach

Contact Info

Product

Resources

About