Multivariate Interaction Analysis of Zea mays L. Genotypes Growth Productivity in Different Environmental Conditions.

Evaluating maize genotypes under different conditions is important for identifying which genotypes combine stability with high yield potential. The aim of this study was to assess stability and the effect of the genotype-environment interaction (GEI) on the grain yield traits of four maize genotypes grown in field trials; one control trial without nitrogen, and three applying different levels of nitrogen (0, 70, 140, and 210 kg ha-1, respectively). Across two growing seasons, both the phenotypic variability and GEI for yield traits over four maize genotypes (P0725, P9889, P9757 and P9074) grown in four different fertilization treatments were studied. The additive main effects and multiplicative interaction (AMMI) models were used to estimate the GEI. The results revealed that genotype and environmental effects, such as the GEI effect, significantly influenced yield, as well as revealing that maize genotypes responded differently to different conditions and fertilization measures. An analysis of the GEI using the IPCA (interaction principal components) analysis method showed the statistical significance of the first source of variation, IPCA1. As the main component, IPCA1 explained 74.6% of GEI variation in maize yield. Genotype G3, with a mean grain yield of 10.6 t ha-1, was found to be the most stable and adaptable to all environments in both seasons, while genotype G1 was found to be unstable, following its specific adaptation to the environments.

Agrobot Lala-An Autonomous Robotic System for Real-Time, In-Field Soil Sampling, and Analysis of Nitrates.

This paper presents an autonomous robotic system, an unmanned ground vehicle (UGV), for in-field soil sampling and analysis of nitrates. Compared to standard methods of soil analysis it has several advantages: each sample is individually analyzed compared to average sample analysis in standard methods; each sample is georeferenced, providing a map for precision base fertilizing; the process is fully autonomous; samples are analyzed in real-time, approximately 30 min per sample; and lightweight for less soil compaction. The robotic system has several modules: commercial robotic platform, anchoring module, sampling module, sample preparation module, sample analysis module, and communication module. The system is augmented with an in-house developed cloud-based platform. This platform uses satellite images, and an artificial intelligence (AI) proprietary algorithm to divide the target field into representative zones for sampling, thus, reducing and optimizing the number and locations of the samples. Based on this, a task is created for the robot to automatically sample at those locations. The user is provided with an in-house developed smartphone app enabling overview and monitoring of the task, changing the positions, removing and adding of the sampling points. The results of the measurements are uploaded to the cloud for further analysis and the creation of prescription maps for variable rate base fertilization.

Behavioural Classification of Cattle Using Neck-Mounted Accelerometer-Equipped Collars.

Monitoring and classification of dairy cattle behaviours is essential for optimising milk yields. Early detection of illness, days before the critical conditions occur, together with automatic detection of the onset of oestrus cycles is crucial for obviating prolonged cattle treatments and improving the pregnancy rates. Accelerometer-based sensor systems are becoming increasingly popular, as they are automatically providing information about key cattle behaviours such as the level of restlessness and the time spent ruminating and eating, proxy measurements that indicate the onset of heat events and overall welfare, at an individual animal level. This paper reports on an approach to the development of algorithms that classify key cattle states based on a systematic dimensionality reduction process through two feature selection techniques. These are based on Mutual Information and Backward Feature Elimination and applied on knowledge-specific and generic time-series extracted from raw accelerometer data. The extracted features are then used to train classification models based on a Hidden Markov Model, Linear Discriminant Analysis and Partial Least Squares Discriminant Analysis. The proposed feature engineering methodology permits model deployment within the computing and memory restrictions imposed by operational settings. The models were based on measurement data from 18 steers, each animal equipped with an accelerometer-based neck-mounted collar and muzzle-mounted halter, the latter providing the truthing data. A total of 42 time-series features were initially extracted and the trade-off between model performance, computational complexity and memory footprint was explored. Results show that the classification model that best balances performance and computation complexity is based on Linear Discriminant Analysis using features selected through Backward Feature Elimination. The final model requires 1.83 ± 1.00 ms to perform feature extraction with 0.05 ± 0.01 ms for inference with an overall balanced accuracy of 0.83.

Classification of Cattle Behaviours Using Neck-Mounted Accelerometer-Equipped Collars and Convolutional Neural Networks.

Monitoring cattle behaviour is core to the early detection of health and welfare issues and to optimise the fertility of large herds. Accelerometer-based sensor systems that provide activity profiles are now used extensively on commercial farms and have evolved to identify behaviours such as the time spent ruminating and eating at an individual animal level. Acquiring this information at scale is central to informing on-farm management decisions. The paper presents the development of a Convolutional Neural Network (CNN) that classifies cattle behavioural states ('rumination', 'eating' and 'other') using data generated from neck-mounted accelerometer collars. During three farm trials in the United Kingdom (Easter Howgate Farm, Edinburgh, UK), 18 steers were monitored to provide raw acceleration measurements, with ground truth data provided by muzzle-mounted pressure sensor halters. A range of neural network architectures are explored and rigorous hyper-parameter searches are performed to optimise the network. The computational complexity and memory footprint of CNN models are not readily compatible with deployment on low-power processors which are both memory and energy constrained. Thus, progressive reductions of the CNN were executed with minimal loss of performance in order to address the practical implementation challenges, defining the trade-off between model performance versus computation complexity and memory footprint to permit deployment on micro-controller architectures. The proposed methodology achieves a compression of 14.30 compared to the unpruned architecture but is nevertheless able to accurately classify cattle behaviours with an overall F1 score of 0.82 for both FP32 and FP16 precision while achieving a reasonable battery lifetime in excess of 5.7 years.

MRI Reconstruction Using Markov Random Field and Total Variation as Composite Prior.

Reconstruction of magnetic resonance images (MRI) benefits from incorporating a priori knowledge about statistical dependencies among the representation coefficients. Recent results demonstrate that modeling intraband dependencies with Markov Random Field (MRF) models enable superior reconstructions compared to inter-scale models. In this paper, we develop a novel reconstruction method, which includes a composite prior based on an MRF model and Total Variation (TV). We use an anisotropic MRF model and propose an original data-driven method for the adaptive estimation of its parameters. From a Bayesian perspective, we define a new position-dependent type of regularization and derive a compact reconstruction algorithm with a novel soft-thresholding rule. Experimental results show the effectiveness of this method compared to the state of the art in the field.

Engineering Meteorological Features to Select Stress Tolerant Hybrids in Maize.

In this study we used meteorological parameters and predictive modelling interpreted by model explanation to develop stress metrics that indicate the presence of drought and heat stress at the specific environment. We started from the extreme temperature and precipitation indices, modified some of them and introduced additional drought indices relevant to the analysis. Based on maize's sensitivity to stress, the growing season was divided into four stages. The features were calculated throughout the growing season and split in two groups, one for the drought and the other for heat stress. Generated meteorological features were combined with soil features and fed to random forest regression model for the yield prediction. Model explanation gave us the contribution of features to yield decrease, from which we estimated total amount of stress at the environments, which represents new environmental index. Using this index we ranked the environments according to the level of stress. More than 2400 hybrids were tested across the environments where they were grown and based on the yield stability they were marked as either tolerant or susceptible to heat, drought or combined heat and drought stress. Presented methodology and results were produced within the Syngenta Crop Challenge 2019.

Portfolio optimization for seed selection in diverse weather scenarios.

The aim of this work was to develop a method for selection of optimal soybean varieties for the American Midwest using data analytics. We extracted the knowledge about 174 varieties from the dataset, which contained information about weather, soil, yield and regional statistical parameters. Next, we predicted the yield of each variety in each of 6,490 observed subregions of the Midwest. Furthermore, yield was predicted for all the possible weather scenarios approximated by 15 historical weather instances contained in the dataset. Using predicted yields and covariance between varieties through different weather scenarios, we performed portfolio optimisation. In this way, for each subregion, we obtained a selection of varieties, that proved superior to others in terms of the amount and stability of yield. According to the rules of Syngenta Crop Challenge, for which this research was conducted, we aggregated the results across all subregions and selected up to five soybean varieties that should be distributed across the network of seed retailers. The work presented in this paper was the winning solution for Syngenta Crop Challenge 2017.

Sparse Recovery in Magnetic Resonance Imaging With a Markov Random Field Prior.

Recent research in compressed sensing of magnetic resonance imaging (CS-MRI) emphasizes the importance of modeling structured sparsity, either in the acquisition or in the reconstruction stages. Subband coefficients of typical images show certain structural patterns, which can be viewed in terms of fixed groups (like wavelet trees) or statistically (certain configurations are more likely than others). Wavelet tree models have already demonstrated excellent performance in MRI recovery from partial data. However, much less attention has been given in CS-MRI to modeling statistically spatial clustering of subband data, although the potentials of such models have been indicated. In this paper, we propose a practical CS-MRI reconstruction algorithm making use of a Markov random field prior model for spatial clustering of subband coefficients and an efficient optimization approach based on proximal splitting. The results demonstrate an improved reconstruction performance compared with both the standard CS-MRI methods and the recent related methods.

Unveiling Spatial Epidemiology of HIV with Mobile Phone Data.

An increasing amount of geo-referenced mobile phone data enables the identification of behavioral patterns, habits and movements of people. With this data, we can extract the knowledge potentially useful for many applications including the one tackled in this study - understanding spatial variation of epidemics. We explored the datasets collected by a cell phone service provider and linked them to spatial HIV prevalence rates estimated from publicly available surveys. For that purpose, 224 features were extracted from mobility and connectivity traces and related to the level of HIV epidemic in 50 Ivory Coast departments. By means of regression models, we evaluated predictive ability of extracted features. Several models predicted HIV prevalence that are highly correlated (>0.7) with actual values. Through contribution analysis we identified key elements that correlate with the rate of infections and could serve as a proxy for epidemic monitoring. Our findings indicate that night connectivity and activity, spatial area covered by users and overall migrations are strongly linked to HIV. By visualizing the communication and mobility flows, we strived to explain the spatial structure of epidemics. We discovered that strong ties and hubs in communication and mobility align with HIV hot spots.

Integrative clustering by nonnegative matrix factorization can reveal coherent functional groups from gene profile data.

Recent developments in molecular biology and techniques for genome-wide data acquisition have resulted in abundance of data to profile genes and predict their function. These datasets may come from diverse sources and it is an open question how to commonly address them and fuse them into a joint prediction model. A prevailing technique to identify groups of related genes that exhibit similar profiles is profile-based clustering. Cluster inference may benefit from consensus across different clustering models. In this paper, we propose a technique that develops separate gene clusters from each of available data sources and then fuses them by means of nonnegative matrix factorization. We use gene profile data on the budding yeast S. cerevisiae to demonstrate that this approach can successfully integrate heterogeneous datasets and yield high-quality clusters that could otherwise not be inferred by simply merging the gene profiles prior to clustering.

Salient motion features for video quality assessment.

Design of algorithms that are able to estimate video quality as perceived by human observers is of interest for a number of applications. Depending on the video content, the artifacts introduced by the coding process can be more or less pronounced and diversely affect the quality of videos, as estimated by humans. While it is well understood that motion affects both human attention and coding quality, this relationship has only recently started gaining attention among the research community, when video quality assessment (VQA) is concerned. In this paper, the effect of calculating several objective measure features, related to video coding artifacts, separately for salient motion and other regions of the frames of the sequence is examined. In addition, we propose a new scheme for quality assessment of coded video streams, which takes into account salient motion. Standardized procedure has been used to calculate the Mean Opinion Score (MOS), based on experiments conducted with a group of non-expert observers viewing standard definition (SD) sequences. MOS measurements were taken for nine different SD sequences, coded using MPEG-2 at five different bit-rates. Eighteen different published approaches related to measuring the amount of coding artifacts objectively on a single-frame basis were implemented. Additional features describing the intensity of salient motion in the frames, as well as the intensity of coding artifacts in the salient motion regions were proposed. Automatic feature selection was performed to determine the subset of features most correlated to video quality. The results show that salient-motion-related features enhance prediction and indicate that the presence of blocking effect artifacts and blurring in the salient regions and variance and intensity of temporal changes in non-salient regions influence the perceived video quality.

Universal impulse noise filter based on genetic programming.

In this paper, we present a novel method for impulse noise filter construction, based on the switching scheme with two cascaded detectors and two corresponding estimators. Genetic programming as a supervised learning algorithm is employed for building two detectors with complementary characteristics. The first detector identifies the majority of noisy pixels. The second detector searches for the remaining noise missed by the first detector, usually hidden in image details or with amplitudes close to its local neighborhood. Both detectors are based on the robust estimators of location and scale-median and MAD. The filter made by the proposed method is capable of effectively suppressing all kinds of impulse noise, in contrast to many existing filters which are specialized only for a particular noise model. In addition, we propose the usage of a new impulse noise model-the mixed impulse noise, which is more realistic and harder to treat than existing impulse noise models. The proposed model is the combination of commonly used noise models: salt-and-pepper and uniform impulse noise models. Simulation results show that the proposed two-stage GP filter produces excellent results and outperforms existing state-of-the-art filters.