Estimation of the nitrogen content of potato plants based on morphological parameters and visible light vegetation indices.

Plant nitrogen content (PNC) is an important indicator to characterize the nitrogen nutrition status of crops, and quickly and efficiently obtaining the PNC information aids in fertilization management and decision-making in modern precision agriculture. This study aimed to explore the potential to improve the accuracy of estimating PNC during critical growth periods of potato by combining the visible light vegetation indices (VIs) and morphological parameters (MPs) obtained from an inexpensive UAV digital camera. First, the visible light VIs and three types of MPs, including the plant height (H), canopy coverage (CC) and canopy volume (CV), were extracted from digital images of the potato tuber formation stage (S1), tuber growth stage (S2), and starch accumulation stage (S3). Then, the correlations of VIs and MPs with the PNC were analyzed for each growth stage, and the performance of VIs and MPs in estimating PNC was explored. Finally, three methods, multiple linear regression (MLR), k-nearest neighbors, and random forest, were used to explore the effect of MPs on the estimation of potato PNC using VIs. The results showed that (i) the values of potato H and CC extracted based on UAV digital images were accurate, and the accuracy of the pre-growth stages was higher than that of the late growth stage. (ii) The estimation of potato PNC by visible light VIs was feasible, but the accuracy required further improvement. (iii) As the growing season progressed, the correlation between MPs and PNC gradually decreased, and it became more difficult to estimate the PNC. (iv) Compared with individual MP, multi-MPs can more accurately reflect the morphological structure of the crop and can further improve the accuracy of estimating PNC. (v) Visible light VIs combined with MPs improved the accuracy of estimating PNC, with the highest accuracy of the models constructed using the MLR method (S1: R 2 = 0.79, RMSE=0.27, NRMSE=8.19%; S2:R 2 = 0.80, RMSE=0.27, NRMSE=8.11%; S3: R 2 = 0.76, RMSE=0.26, NRMSE=8.63%). The results showed that the combination of visible light VIs and morphological information obtained by a UAV digital camera could provide a feasible method for monitoring crop growth and plant nitrogen status.

Estimation of potato above-ground biomass based on unmanned aerial vehicle red-green-blue images with different texture features and crop height.

Obtaining crop above-ground biomass (AGB) information quickly and accurately is beneficial to farmland production management and the optimization of planting patterns. Many studies have confirmed that, due to canopy spectral saturation, AGB is underestimated in the multi-growth period of crops when using only optical vegetation indices. To solve this problem, this study obtains textures and crop height directly from ultrahigh-ground-resolution (GDS) red-green-blue (RGB) images to estimate the potato AGB in three key growth periods. Textures include a grayscale co-occurrence matrix texture (GLCM) and a Gabor wavelet texture. GLCM-based textures were extracted from seven-GDS (1, 5, 10, 30, 40, 50, and 60 cm) RGB images. Gabor-based textures were obtained from magnitude images on five scales (scales 1-5, labeled S1-S5, respectively). Potato crop height was extracted based on the generated crop height model. Finally, to estimate potato AGB, we used (i) GLCM-based textures from different GDS and their combinations, (ii) Gabor-based textures from different scales and their combinations, (iii) all GLCM-based textures combined with crop height, (iv) all Gabor-based textures combined with crop height, and (v) two types of textures combined with crop height by least-squares support vector machine (LSSVM), extreme learning machine, and partial least squares regression techniques. The results show that (i) potato crop height and AGB first increase and then decrease over the growth period; (ii) GDS and scales mainly affect the correlation between GLCM- and Gabor-based textures and AGB; (iii) to estimate AGB, GLCM-based textures of GDS1 and GDS30 work best when the GDS is between 1 and 5 cm and 10 and 60 cm, respectively (however, estimating potato AGB based on Gabor-based textures gradually deteriorates as the Gabor convolution kernel scale increases); (iv) the AGB estimation based on a single-type texture is not as good as estimates based on multi-resolution GLCM-based and multiscale Gabor-based textures (with the latter being the best); (v) different forms of textures combined with crop height using the LSSVM technique improved by 22.97, 14.63, 9.74, and 8.18% (normalized root mean square error) compared with using only all GLCM-based textures, all Gabor-based textures, the former combined with crop height, and the latter combined with crop height, respectively. Therefore, different forms of texture features obtained from RGB images acquired from unmanned aerial vehicles and combined with crop height improve the accuracy of potato AGB estimates under high coverage.

Method for accurate multi-growth-stage estimation of fractional vegetation cover using unmanned aerial vehicle remote sensing.

BACKGROUND: Fractional vegetation cover (FVC) is an important parameter for evaluating crop-growth status. Optical remote-sensing techniques combined with the pixel dichotomy model (PDM) are widely used to estimate cropland FVC with medium to high spatial resolution on the ground. However, PDM-based FVC estimation is limited by effects stemming from the variation of crop canopy chlorophyll content (CCC). To overcome this difficulty, we propose herein a "fan-shaped method" (FSM) that uses a CCC spectral index (SI) and a vegetation SI to create a two-dimensional scatter map in which the three vertices represent high-CCC vegetation, low-CCC vegetation, and bare soil. The FVC at each pixel is determined based on the spatial location of the pixel in the two-dimensional scatter map, which mitigates the effects of CCC on the PDM. To evaluate the accuracy of FSM estimates of the FVC, we analyze the spectra obtained from (a) the PROSAIL model and (b) a spectrometer mounted on an unmanned aerial vehicle platform. Specifically, we use both the proposed FSM and traditional remote-sensing FVC-estimation methods (both linear and nonlinear regression and PDM) to estimate soybean FVC. RESULTS: Field soybean CCC measurements indicate that (a) the soybean CCC increases continuously from the flowering growth stage to the later-podding growth stage, and then decreases with increasing crop growth stages, (b) the coefficient of variation of soybean CCC is very large in later growth stages (31.58-35.77%) and over all growth stages (26.14%). FVC samples with low CCC are underestimated by the PDM. Linear and nonlinear regression underestimates (overestimates) FVC samples with low (high) CCC. The proposed FSM depends less on CCC and is thus a robust method that can be used for multi-stage FVC estimation of crops with strongly varying CCC. CONCLUSIONS: Estimates and maps of FVC based on the later growth stages and on multiple growth stages should consider the variation of crop CCC. FSM can mitigates the effect of CCC by conducting a PDM at each CCC level. The FSM is a robust method that can be used to estimate FVC based on multiple growth stages where crop CCC varies greatly.

A robust spectral angle index for remotely assessing soybean canopy chlorophyll content in different growing stages.

BACKGROUND: Timely and accurate estimates of canopy chlorophyll (Chl) a and b content are crucial for crop growth monitoring and agricultural management. Crop canopy reflectance depends on many factors, which can be divided into the following categories: (i) leaf effects (e.g., leaf pigments), (ii) canopy effects (e.g., Leaf Area Index [LAI]), and (iii) soil background reflectance (e.g., soil reflectance). The estimation of leaf variables, such as Chl contents, from reflectance at the canopy scale is usually less accurate than that at the leaf scale. In this study, we propose a Visible and Near-infrared (NIR) Angle Index (VNAI) to estimate the Chl content of soybean canopy, and soybean canopy Chl maps are produced using visible and NIR unmanned aerial vehicle (UAV) remote sensing images. The VNAI is insensitive to LAI and can be used for the multi-stage estimation of crop canopy Chl content. RESULTS: Eleven previously used vegetation indices (VIs) (e.g., Pigment-specific Normalized Difference Index) were selected for performance comparison. The results showed that (i) most previously used Chl VIs were significantly correlated with LAI, and the proposed VNAI was more sensitive to Chl content than LAI; (ii) the VNAI-based estimates of Chl content were more accurate than those based on the other investigated VIs using (1) simulated, (2) real (field), and (3) real (UAV) datasets. CONCLUSIONS: Most previously used Chl VIs were significantly correlated with LAI whereas the proposed VNAI was more sensitive to Chl content than to LAI, indicating that the VNAI may be more strongly correlated with Chl content than these previously used VIs. Multi-stage estimations of the Chl content of cropland obtained using the VNAI and broadband remote sensing images may help to obtain Chl maps with high temporal and spatial resolution.

A Novel Greenhouse-Based System for the Detection and Plumpness Assessment of Strawberry Using an Improved Deep Learning Technique.

The automated harvesting of strawberry brings benefits such as reduced labor costs, sustainability, increased productivity, less waste, and improved use of natural resources. The accurate detection of strawberries in a greenhouse can be used to assist in the effective recognition and location of strawberries for the process of strawberry collection. Furthermore, being able to detect and characterize strawberries based on field images is an essential component in the breeding pipeline for the selection of high-yield varieties. The existing manual examination method is error-prone and time-consuming, which makes mechanized harvesting difficult. In this work, we propose a robust architecture, named "improved Faster-RCNN," to detect strawberries in ground-level RGB images captured by a self-developed "Large Scene Camera System." The purpose of this research is to develop a fully automatic detection and plumpness grading system for living plants in field conditions which does not require any prior information about targets. The experimental results show that the proposed method obtained an average fruit extraction accuracy of more than 86%, which is higher than that obtained using three other methods. This demonstrates that image processing combined with the introduced novel deep learning architecture is highly feasible for counting the number of, and identifying the quality of, strawberries from ground-level images. Additionally, this work shows that deep learning techniques can serve as invaluable tools in larger field investigation frameworks, specifically for applications involving plant phenotyping.

A Monitoring System for the Segmentation and Grading of Broccoli Head Based on Deep Learning and Neural Networks.

Achieving the non-contact and non-destructive observation of broccoli head is the key step to realize the acquisition of high-throughput phenotyping information of broccoli. However, the rapid segmentation and grading of broccoli head remains difficult in many parts of the world due to low equipment development level. In this paper, we combined an advanced computer vision technique with a deep learning architecture to allow the acquisition of real-time and accurate information about broccoli head. By constructing a private image dataset with 100s of broccoli-head images (acquired using a self-developed imaging system) under controlled conditions, a deep convolutional neural network named "Improved ResNet" was trained to extract the broccoli pixels from the background. Then, a yield estimation model was built based on the number of extracted pixels and the corresponding pixel weight value. Additionally, the Particle Swarm Optimization Algorithm (PSOA) and the Otsu method were applied to grade the quality of each broccoli head according to our new standard. The trained model achieved an Accuracy of 0.896 on the test set for broccoli head segmentation, demonstrating the feasibility of this approach. When testing the model on a set of images with different light intensities or with some noise, the model still achieved satisfactory results. Overall, our approach of training a deep learning model using low-cost imaging devices represents a means to improve broccoli breeding and vegetable trade.

Estimation of Crop Growth Parameters Using UAV-Based Hyperspectral Remote Sensing Data.

Above-ground biomass (AGB) and the leaf area index (LAI) are important indicators for the assessment of crop growth, and are therefore important for agricultural management. Although improvements have been made in the monitoring of crop growth parameters using ground- and satellite-based sensors, the application of these technologies is limited by imaging difficulties, complex data processing, and low spatial resolution. Therefore, this study evaluated the use of hyperspectral indices, red-edge parameters, and their combination to estimate and map the distributions of AGB and LAI for various growth stages of winter wheat. A hyperspectral sensor mounted on an unmanned aerial vehicle was used to obtain vegetation indices and red-edge parameters, and stepwise regression (SWR) and partial least squares regression (PLSR) methods were used to accurately estimate the AGB and LAI based on these vegetation indices, red-edge parameters, and their combination. The results show that: (i) most of the studied vegetation indices and red-edge parameters are significantly highly correlated with AGB and LAI; (ii) overall, the correlations between vegetation indices and AGB and LAI, respectively, are stronger than those between red-edge parameters and AGB and LAI, respectively; (iii) Compared with the estimations using only vegetation indices or red-edge parameters, the estimation of AGB and LAI using a combination of vegetation indices and red-edge parameters is more accurate; and (iv) The estimations of AGB and LAI obtained using the PLSR method are superior to those obtained using the SWR method. Therefore, combining vegetation indices with red-edge parameters and using the PLSR method can improve the estimation of AGB and LAI.

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

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.

Wheat Ears Counting in Field Conditions Based on Multi-Feature Optimization and TWSVM.

The number of wheat ears in the field is very important data for predicting crop growth and estimating crop yield and as such is receiving ever-increasing research attention. To obtain such data, we propose a novel algorithm that uses computer vision to accurately recognize wheat ears in a digital image. First, red-green-blue images acquired by a manned ground vehicle are selected based on light intensity to ensure that this method is robust with respect to light intensity. Next, the selected images are cut to ensure that the target can be identified in the remaining parts. The simple linear iterative clustering method, which is based on superpixel theory, is then used to generate a patch from the selected images. After manually labeling each patch, they are divided into two categories: wheat ears and background. The color feature "Color Coherence Vectors," the texture feature "Gray Level Co-Occurrence Matrix," and a special image feature "Edge Histogram Descriptor" are then exacted from these patches to generate a high-dimensional matrix called the "feature matrix." Because each feature plays a different role in the classification process, a feature-weighting fusion based on kernel principal component analysis is used to redistribute the feature weights. Finally, a twin-support-vector-machine segmentation (TWSVM-Seg) model is trained to understand the differences between the two types of patches through the features, and the TWSVM-Seg model finally achieves the correct classification of each pixel from the testing sample and outputs the results in the form of binary image. This process thus segments the image. Next, we use a statistical function in Matlab to get the exact a precise number of ears. To verify these statistical numerical results, we compare them with field measurements of the wheat plots. The result of applying the proposed algorithm to ground-shooting image data sets correlates strongly (with a precision of 0.79-0.82) with the data obtained by manual counting. An average running time of 0.1 s is required to successfully extract the correct number of ears from the background, which shows that the proposed algorithm is computationally efficient. These results indicate that the proposed method provides accurate phenotypic data on wheat seedlings.