1722 research outputs found
Sort by
A comprehensive review of 3D convolutional neural network-based classification techniques of diseased and defective crops using non-UAV-based hyperspectral images
Get PDFHyperspectral imaging (HSI) is a non-destructive and contactless technology that provides valuable information about the structure and composition of an object. It has the ability to capture detailed information about the chemical and physical properties of agricultural crops. Due to its wide spectral range, compared with multispectral-or RGB-based imaging methods, HSI can be a more effective tool for monitoring crop health and productivity. With the advent of this imaging tool in agrotechnology, researchers can more accurately address issues related to the detection of diseased and defective crops in the agriculture industry. This allows to implement the most suitable and accurate farming solutions, such as irrigation and fertilization, before crops enter a damaged and difficult-to-recover phase of growth in the field. While HSI provides valuable insights into the object under investigation, the limited number of HSI datasets for crop evaluation presently poses a bottleneck. Dealing with the curse of dimensionality presents another challenge due to the abundance of spectral and spatial information in each hyperspectral cube. State-of-the-art methods based on 1D and 2D convolutional neural networks (CNNs) struggle to efficiently extract spectral and spatial information. On the other hand, 3D-CNN-based models have shown significant promise in achieving better classification and detection results by leveraging spectral and spatial features simultaneously. Despite the apparent benefits of 3D-CNN-based models, their usage for classification purposes in this area of research has remained limited. This paper seeks to address this gap by reviewing 3D-CNN-based architectures and the typical deep learning pipeline, including preprocessing and visualization of results, for the classification of hyperspectral images of diseased and defective crops. Furthermore, we discuss open research areas and challenges when utilizing 3D-CNNs with HSI data."This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors."https://www.sciencedirect.com/science/article/pii/S277237552300145
Chalcogen-substituted carbenes: a density functional study of structure, stability, and donor ability
No full textChalcogen-substituted carbenes are examined computationally using density functional theory. Several approaches are used to assess the stability and reactivity of chalcogenazol-2-ylidene carbenes (NEHCs; E = O, S, Se, Te). The known unsaturated species 1,3-dimethylimidazol-2-ylidene is studied at the same level of theory as the NEHC molecules, as a reference. Electronic structures, stability towards dimerization, and ligand properties are discussed. The results highlight the NEHCs as potentially valuable ancillary ligands for stabilizing low-valent metals or paramagnetic main group molecules. A simple, effective computational method for evaluating s donor ability and p acidity of carbenes is presented.NSERC (grant No. RGPIN-2019-06725),
Digital Research Alliance Canadahttps://pubs.rsc.org/en/content/articlelanding/2023/ra/d3ra03324
Emotion Recognition with Asymmetry Features of EEG Signals
Get PDFCurrently the study of affective computing (AC) includes a focus on researching emotion regulation and recognition. Recent studies in this field have utilized deep learning architectures to enhance emotion recognition from EEG signals. An alternative approach to deep learning is to use feature engineering to extract relevant features to train supervised machine learning models. Current theories in the neuroscience field can guide this feature engineering process. Neuroscientists have suggested various models to clarify how emotions are processed. One of these models suggests that positive emotions are processed in the left hemisphere, while negative emotions are processed in the right hemisphere. This emotional processing model has inspired previous studies to propose asymmetrical features to predict emotions. However, none of these studies have statistically evaluated whether the inclusion of asymmetrical features could yield benefits such as increased accuracy or reduced training time. To address that direction, this research presents both statistical evaluations for emotion regulation and a comparable model for emotion recognition. The outcomes show that brain hemispheres and frequency bands participate differently in processing emotions and observed the presence of the two asymmetry emotion processing models but in different frequency ranges. Also, the results from this study imply that by using asymmetry EEG, emotion recognition approaches can use fewer features without significantly compromising performance.Master of Science in Applied Computer Scienc
Securing Federated Learning Model Aggregation Against Poisoning Attacks via Credit-Based Client Selection
Get PDFFederated Learning (FL) has emerged as a revolutionary paradigm in the field of machine learning, enabling multiple participants to collaboratively train models without compromising the privacy of their individual training data. However, the distributed and decentralized nature of FL also exposes it to a diverse array of poisoning attacks, wherein adversaries inject malicious updates to compromise the integrity and accuracy of the global model.In this thesis, we embark on a critical exploration of defense strategies against poisoning attacks in FL. Our primary focus lies in proposing and evaluating a robust defense mechanism, aptly named Credit-Based Client Selection (CBCS). Leveraging a credit-based system, CBCS judiciously assigns credit scores to participating clients based on the accuracy and consistency of their historical model updates. By selectively incorporating reliable clients with higher credit scores into the model aggregation process, while subjecting low-credit clients to thorough scrutiny or exclusion, CBCS fortifies the defense against adversarial disruptions. To further enhance our research comprehensiveness, we extend our evaluation to other scenarios that can be explored, such as normal conditions. We carefully assess the efficacy of these strategies across various FL settings.Through an extensive series of experiments conducted on non-iid image classification datasets, we rigorously evaluate the performance of the CBCS defense mechanism. The results show that CBCS effectively identifies and excludes adversarial clients, maintaining model accuracy and data confidentiality in federated learning. The outcomes of our research underscore the profound impact of robust defense strategies on securing federated learning and their pivotal role in advancing collaborative and privacy-preserving machine learning applications.The proposed CBCS defense mechanism illuminates new avenues for enhancing the resilience and security of federated learning systems in the face of adversarial threats.As the world continues to embrace decentralized and privacy-focused learning approaches, our research contributes significantly to the safe and trustworthy deployment of federated learning across diverse domains.Master of Science in Applied Computer Scienc
Coyotes, cattle, and native prey in southwest Saskatchewan
Get PDFNative grasslands are home to a disproportionate share of species at risk, and typically consist of a mosaic of ranchland and protected parkland. Livestock carcasses can attract coyotes and potentially subsidize the coyote population or increase depredation, which is a concern to ranchers in southwest Saskatchewan. A subsidized predator population may increase pressure on native prey species through apparent competition. In this thesis, I investigated the relationship between coyotes, cattle and native prey. In the second chapter, I used molecular methods to test how commonly coyotes consumed cattle and species at risk, and how geographic factors affected the presence of cattle versus deer in coyote diet. Deer and cattle were the most common food items. Scat containing cattle was typically found closer to a boneyard and the bison enclosure, whereas scat containing deer was typically further from a boneyard and the bison enclosure. Different individual coyotes may be consuming cattle versus deer and coyotes consuming cattle may show different travel behaviour than coyotes consuming native prey. However, I found no evidence that coyotes pose a direct threat to species at risk during the winter. In the third chapter, I observed coyotes during summer to test whether coyotes obtained direct and/or indirect benefits from cattle pastures, and how cows responded to the presence of coyotes. Coyotes hunted native prey and specifically ground squirrels more commonly than cattle, showing that they obtained indirect benefits from the use of cattle pastures. Cows responded to coyotes defensively, and although observations of coyotes approaching individual calves, rushing cow-calf herds, or harassing females for afterbirth were uncommon, these observations, combined with the coyotes’ scavenging from cattle carcasses, indicate that coyotes also benefit directly by consuming cattle or cattle by-products. Further work identifying individual coyotes would help to determine what proportion of the population is being subsidized by cattle and factors that might predispose individual coyotes to depredation.Natural Sciences and Engineering Research Council of CanadaMaster of Science in Bioscience, Technology, and Public Polic
Exploring Hyperspectral Imaging and 3D Convolutional Neural Network for Stress Classification in Plants
Get PDFHyperspectral imaging (HSI) has emerged as a transformative technology in imaging, characterized by its ability to capture a wide spectrum of light, including wavelengths beyond the visible range. This approach significantly differs from traditional imaging methods such as RGB imaging, which uses three color channels, and multispectral imaging, which captures several discrete spectral bands. Through this approach, HSI offers detailed spectral signatures for each pixel, facilitating a more nuanced analysis of the imaged subjects. This capability is particularly beneficial in applications like agricultural practices, where it can detect changes in physiological and structural characteristics of crops. Moreover, the ability of HSI to monitor these changes over time is advantageous for observing how subjects respond to different environmental conditions or treatments. However, the high-dimensional nature of hyperspectral data presents challenges in data processing and feature extraction. Traditional machine learning algorithms often struggle to handle such complexity. This is where 3D Convolutional Neural Networks (CNNs) become valuable. Unlike 1D-CNNs, which extract features from spectral dimensions, and 2D-CNNs, which focus on spatial dimensions, 3D CNNs have the capability to process data across both spectral and spatial dimensions. This makes them adept at extracting complex features from hyperspectral data. In this thesis, we explored the potency of HSI combined with 3D-CNN in agriculture domain where plant health and vitality are paramount. To evaluate this, we subjected lettuce plants to varying stress levels to assess the performance of this method in classifying the stressed lettuce at the early stages of growth into their respective stress-level groups. For this study, we created a dataset comprising 88 hyperspectral image samples of stressed lettuce. Utilizing Bayesian optimization, we developed 350 distinct 3D-CNN models to assess the method. The top-performing model achieved a 75.00\% test accuracy. Additionally, we addressed the challenge of generating valid 3D-CNN models in the Keras Tuner library through meticulous hyperparameter configuration. Our investigation also extends to the role of individual channels and channel groups within the color and near-infrared spectrum in predicting results for each stress-level group. We observed that the red and green spectra have a higher influence on the prediction results. Furthermore, we conducted a comprehensive review of 3D-CNN-based classification techniques for diseased and defective crops using non-UAV-based hyperspectral images.MITACSMaster of Science in Applied Computer Scienc
Resolving Stock Structure of Sauger (Sander canadensis) in Manitoba, Canada using Biometric, Isotopic, and Genetic Approaches
Get PDFMany sauger (Sander canadensis) populations in Manitoba have declined in numbers and biomass. Fisheries managers have proposed a province-wide sauger management plan to protect and restore sauger populations, but they are uncertain how sauger populations should be defined and to what extent they may interact. In this thesis, I used a multifaceted approach to resolve population structure and identify migratory corridors of sauger in Manitoba. First, I mined biometric data from several long-term monitoring datasets to calculate life history indices for sauger stocks across 29 waterbodies. Sauger growth generally decreased and the age at 50% maturity increased among lakes of increasing latitude. This trend was also observed within Lake Winnipeg, yet the length at 50% maturity remained constant. Sauger grew exceptionally fast in Lake Manitoba and Lake Winnipegosis and matured at an early age. Next, I performed a stable isotope analysis (13C and 15N) of sauger tissue to investigate contemporary sauger migration throughout the Lake Winnipeg watershed. Sauger from Lake Winnipeg, Lake Manitoba, and Lac du Bonnet occupied distinct isotopic niches, and I identified several possible migrants from Lake Manitoba and the Winnipeg River in Lake Winnipeg. Finally, I used microsatellites to assess the genetic health and structure of sauger stocks across Manitoba. Genetic diversity within sample populations was moderate to high, and incidence of inbreeding and hybridization with walleye (Sander vitreus) was low. I identified four broad genetic sauger stocks: Lake Winnipeg; Lake Manitoba and Lake Winnipegosis; the Red and Assiniboine Rivers; and the Churchill and Saskatchewan Rivers. Gene flow between Lake Winnipeg and Lake Manitoba stocks is minimal. These findings will assist managers in defining stock management units and optimizing management efforts for sauger populations in Manitoba.Manitoba Fish and Wildlife Enhancement Fund (#FES19-010); Fish Futures (Dr. Ken Stewart Memorial Scholarship, 2021).Master of Science in Bioscience, Technology, and Public Polic
Documentation on traditional Indigenous Material Culture in Books for Young Readers
No full textThe Six Seasons Project of Asiniskaw Īthiniwak project is supported in part by funding from the Social Sciences and Humanities Research Council