A Real-time Low-cost Artificial Intelligence System for Autonomous Spraying in Palm Plantations

In precision crop protection, (target-orientated) object detection in image processing can help navigate Unmanned Aerial Vehicles (UAV, crop protection drones) to the right place to apply the pesticide. Unnecessary application of non-target areas could be avoided. Deep learning algorithms dominantly use in modern computer vision tasks which require high computing time, memory footprint, and power consumption. Based on the Edge Artificial Intelligence, we investigate the main three paths that lead to dealing with this problem, including hardware accelerators, efficient algorithms, and model compression. Finally, we integrate them and propose a solution based on a light deep neural network (DNN), called Ag-YOLO, which can make the crop protection UAV have the ability to target detection and autonomous operation. This solution is restricted in size, cost, flexible, fast, and energy-effective. The hardware is only 18 grams in weight and 1.5 watts in energy consumption, and the developed DNN model needs only 838 kilobytes of disc space. We tested the developed hardware and software in comparison to the tiny version of the state-of-art YOLOv3 framework, known as YOLOv3-Tiny to detect individual palm in a plantation. An average F1 score of 0.9205 at the speed of 36.5 frames per second (in comparison to similar accuracy at 18 frames per second and 8.66 megabytes of the YOLOv3-Tiny algorithm) was reached. This developed detection system is easily plugged into any machines already purchased as long as the machines have USB ports and run Linux Operating System.

Radar-based Materials Classification Using Deep Wavelet Scattering Transform: A Comparison of Centimeter vs. Millimeter Wave Units

Radar-based materials detection received significant attention in recent years for its potential inclusion in consumer and industrial applications like object recognition for grasping and manufacturing quality assurance and control. Several radar publications were developed for material classification under controlled settings with specific materials' properties and shapes. Recent literature has challenged the earlier findings on radars-based materials classification claiming that earlier solutions are not easily scaled to industrial applications due to a variety of real-world issues. Published experiments on the impact of these factors on the robustness of the extracted radar-based traditional features have already demonstrated that the application of deep neural networks can mitigate, to some extent, the impact to produce a viable solution. However, previous studies lacked an investigation of the usefulness of lower frequency radar units, specifically <10GHz, against the higher range units around and above 60GHz. This research considers two radar units with different frequency ranges: Walabot-3D (6.3-8 GHz) cm-wave and IMAGEVK-74 (62-69 GHz) mm-wave imaging units by Vayyar Imaging. A comparison is presented on the applicability of each unit for material classification. This work extends upon previous efforts, by applying deep wavelet scattering transform for the identification of different materials based on the reflected signals. In the wavelet scattering feature extractor, data is propagated through a series of wavelet transforms, nonlinearities, and averaging to produce low-variance representations of the reflected radar signals. This work is unique in comparison of the radar units and algorithms in material classification and includes real-time demonstrations that show strong performance by both units, with increased robustness offered by the cm-wave radar unit.

Choosing an Appropriate Platform and Workflow for Processing Camera Trap Data using Artificial Intelligence

Camera traps have transformed how ecologists study wildlife species distributions, activity patterns, and interspecific interactions. Although camera traps provide a cost-effective method for monitoring species, the time required for data processing can limit survey efficiency. Thus, the potential of Artificial Intelligence (AI), specifically Deep Learning (DL), to process camera-trap data has gained considerable attention. Using DL for these applications involves training algorithms, such as Convolutional Neural Networks (CNNs), to automatically detect objects and classify species. To overcome technical challenges associated with training CNNs, several research communities have recently developed platforms that incorporate DL in easy-to-use interfaces. We review key characteristics of four AI-powered platforms -- Wildlife Insights (WI), MegaDetector (MD), Machine Learning for Wildlife Image Classification (MLWIC2), and Conservation AI -- including data management tools and AI features. We also provide R code in an open-source GitBook, to demonstrate how users can evaluate model performance, and incorporate AI output in semi-automated workflows. We found that species classifications from WI and MLWIC2 generally had low recall values (animals that were present in the images often were not classified to the correct species). Yet, the precision of WI and MLWIC2 classifications for some species was high (i.e., when classifications were made, they were generally accurate). MD, which classifies images using broader categories (e.g., "blank" or "animal"), also performed well. Thus, we conclude that, although species classifiers were not accurate enough to automate image processing, DL could be used to improve efficiencies by accepting classifications with high confidence values for certain species or by filtering images containing blanks.

Multivariate time-series modeling with generative neural networks

Generative moment matching networks (GMMNs) are introduced as dependence models for the joint innovation distribution of multivariate time series (MTS). Following the popular copula-GARCH approach for modeling dependent MTS data, a framework allowing us to take an alternative GMMN-GARCH approach is presented. First, ARMA-GARCH models are utilized to capture the serial dependence within each univariate marginal time series. Second, if the number of marginal time series is large, principal component analysis (PCA) is used as a dimension-reduction step. Last, the remaining cross-sectional dependence is modeled via a GMMN, our main contribution. GMMNs are highly flexible and easy to simulate from, which is a major advantage over the copula-GARCH approach. Applications involving yield curve modeling and the analysis of foreign exchange rate returns are presented to demonstrate the utility of our approach, especially in terms of producing better empirical predictive distributions and making better probabilistic forecasts. All results are reproducible with the demo GMMN_MTS_paper of the R package gnn.

Unsupervised Sparse Unmixing of Atmospheric Trace Gases from Hyperspectral Satellite Data

In this letter, a new approach for the retrieval of the vertical column concentrations of trace gases from hyperspectral satellite observations, is proposed. The main idea is to perform a linear spectral unmixing by estimating the abundances of trace gases spectral signatures in each mixed pixel collected by an imaging spectrometer in the ultraviolet region. To this aim, the sparse nature of the measurements is brought to light and the compressive sensing paradigm is applied to estimate the concentrations of the gases' endemembers given by an a priori wide spectral library, including reference cross sections measured at different temperatures and pressures at the same time. The proposed approach has been experimentally assessed using both simulated and real hyperspectral dataset. Specifically, the experimental analysis relies on the retrieval of sulfur dioxide during volcanic emissions using data collected by the TROPOspheric Monitoring Instrument. To validate the procedure, we also compare the obtained results with the sulfur dioxide total column product based on the differential optical absorption spectroscopy technique and the retrieved concentrations estimated using the blind source separation.

Scalable and Decentralized Algorithms for Anomaly Detection via Learning-Based Controlled Sensing

We address the problem of sequentially selecting and observing processes from a given set to find the anomalies among them. The decision-maker observes a subset of the processes at any given time instant and obtains a noisy binary indicator of whether or not the corresponding process is anomalous. In this setting, we develop an anomaly detection algorithm that chooses the processes to be observed at a given time instant, decides when to stop taking observations, and declares the decision on anomalous processes. The objective of the detection algorithm is to identify the anomalies with an accuracy exceeding the desired value while minimizing the delay in decision making. We devise a centralized algorithm where the processes are jointly selected by a common agent as well as a decentralized algorithm where the decision of whether to select a process is made independently for each process. Our algorithms rely on a Markov decision process defined using the marginal probability of each process being normal or anomalous, conditioned on the observations. We implement the detection algorithms using the deep actor-critic reinforcement learning framework. Unlike prior work on this topic that has exponential complexity in the number of processes, our algorithms have computational and memory requirements that are both polynomial in the number of processes. We demonstrate the efficacy of these algorithms using numerical experiments by comparing them with state-of-the-art methods.

Broadband Radar Invisibility with Time-Dependent Metasurfaces

Concealing objects from interrogation has been a primary objective since the integration of radars into surveillance systems. Metamaterial-based invisibility cloaking, which was considered a promising solution, did not yet succeed in delivering reliable performance against real radar systems, mainly due to its narrow operational bandwidth. Here we propose an approach, which addresses the issue from a signal-processing standpoint and, as a result, is capable of coping with the vast majority of unclassified radar systems by exploiting vulnerabilities in their design. In particular, we demonstrate complete concealment of a 0.25 square meter moving metal plate from an investigating radar system, operating in a broad frequency range approaching 20% bandwidth around the carrier of 1.5GHz. The key element of the radar countermeasure is a temporally modulated coating. This auxiliary structure is designed to dynamically and controllably adjust the reflected phase of the impinging radar signal, which acquires a user-defined Doppler shift. A special case of interest is imposing a frequency shift that compensates for the real Doppler signatures originating from the motion of the target. In this case the radar will consider the target static, even though it is moving. As a result, the reflected echo will be discarded by the clutter removal filter, which is an inherent part of any modern radar system that is designed to operate in real conditions. This signal-processing loophole allows rendering the target invisible to the radar even though it scatters electromagnetic radiation. This application claims the benefit of patent number 273995, filed on April 16 2020.

Explaining Naive Bayes and Other Linear Classifiers with Polynomial Time and Delay

Recent work proposed the computation of so-called PI-explanations of Naive Bayes Classifiers (NBCs). PI-explanations are subset-minimal sets of feature-value pairs that are sufficient for the prediction, and have been computed with state-of-the-art exact algorithms that are worst-case exponential in time and space. In contrast, we show that the computation of one PI-explanation for an NBC can be achieved in log-linear time, and that the same result also applies to the more general class of linear classifiers. Furthermore, we show that the enumeration of PI-explanations can be obtained with polynomial delay. Experimental results demonstrate the performance gains of the new algorithms when compared with earlier work. The experimental results also investigate ways to measure the quality of heuristic explanations

Scribble-based Boundary-aware Network for Weakly Supervised Salient Object Detection in Remote Sensing Images

Existing CNNs-based salient object detection (SOD) heavily depends on the large-scale pixel-level annotations, which is labor-intensive, time-consuming, and expensive. By contrast, the sparse annotations become appealing to the salient object detection community. However, few efforts are devoted to learning salient object detection from sparse annotations, especially in the remote sensing field. In addition, the sparse annotation usually contains scanty information, which makes it challenging to train a well-performing model, resulting in its performance largely lagging behind the fully-supervised models. Although some SOD methods adopt some prior cues to improve the detection performance, they usually lack targeted discrimination of object boundaries and thus provide saliency maps with poor boundary localization. To this end, in this paper, we propose a novel weakly-supervised salient object detection framework to predict the saliency of remote sensing images from sparse scribble annotations. To implement it, we first construct the scribble-based remote sensing saliency dataset by relabelling an existing large-scale SOD dataset with scribbles, namely S-EOR dataset. After that, we present a novel scribble-based boundary-aware network (SBA-Net) for remote sensing salient object detection. Specifically, we design a boundary-aware module (BAM) to explore the object boundary semantics, which is explicitly supervised by the high-confidence object boundary (pseudo) labels generated by the boundary label generation (BLG) module, forcing the model to learn features that highlight the object structure and thus boosting the boundary localization of objects. Then, the boundary semantics are integrated with high-level features to guide the salient object detection under the supervision of scribble labels.

Crop mapping from image time series: deep learning with multi-scale label hierarchies

The aim of this paper is to map agricultural crops by classifying satellite image time series. Domain experts in agriculture work with crop type labels that are organised in a hierarchical tree structure, where coarse classes (like orchards) are subdivided into finer ones (like apples, pears, vines, etc.). We develop a crop classification method that exploits this expert knowledge and significantly improves the mapping of rare crop types. The three-level label hierarchy is encoded in a convolutional, recurrent neural network (convRNN), such that for each pixel the model predicts three labels at different level of granularity. This end-to-end trainable, hierarchical network architecture allows the model to learn joint feature representations of rare classes (e.g., apples, pears) at a coarser level (e.g., orchard), thereby boosting classification performance at the fine-grained level. Additionally, labelling at different granularity also makes it possible to adjust the output according to the classification scores; as coarser labels with high confidence are sometimes more useful for agricultural practice than fine-grained but very uncertain labels. We validate the proposed method on a new, large dataset that we make public. ZueriCrop covers an area of 50 km x 48 km in the Swiss cantons of Zurich and Thurgau with a total of 116'000 individual fields spanning 48 crop classes, and 28,000 (multi-temporal) image patches from Sentinel-2. We compare our proposed hierarchical convRNN model with several baselines, including methods designed for imbalanced class distributions. The hierarchical approach performs superior by at least 9.9 percentage points in F1-score.