arXiv subjects

Bronson Philippa

Publications and source records attributed to Bronson Philippa.

14 recordsLinked to original sources

arXiv2024

Semi-Supervised Weed Detection for Rapid Deployment and Enhanced Efficiency

Weeds present a significant challenge in agriculture, causing yield loss and requiring expensive control measures. Automatic weed detection using computer vision and deep learning offers a promising solution. However, conventional deep learning methods often require large amounts of labelled training data, which can be costly and time-consuming to acquire. This paper introduces a novel method for semi-supervised weed detection, comprising two main components. Firstly, a multi-scale feature representation technique is employed to capture distinctive weed features across different scales. Secondly, we propose an adaptive pseudo-label assignment strategy, leveraging a small set of labelled images during training. This strategy dynamically assigns confidence scores to pseudo-labels generated from unlabeled data. Additionally, our approach integrates epoch-corresponding and mixed pseudo-labels to further enhance the learning process. Experimental results on the COCO dataset and five prominent weed datasets -- CottonWeedDet12, CropAndWeed, Palmer amaranth, RadishWheat, and RoboWeedMap -- illustrate that our method achieves state-of-the-art performance in weed detection, even with significantly less labelled data compared to existing techniques. This approach holds the potential to alleviate the labelling burden and enhance the feasibility and deployment speed of deep learning for weed detection in real-world agricultural scenarios.

cs.CV↗

arXiv2024

FieldNet: Efficient Real-Time Shadow Removal for Enhanced Vision in Field Robotics

Shadows significantly hinder computer vision tasks in outdoor environments, particularly in field robotics, where varying lighting conditions complicate object detection and localisation. We present FieldNet, a novel deep learning framework for real-time shadow removal, optimised for resource-constrained hardware. FieldNet introduces a probabilistic enhancement module and a novel loss function to address challenges of inconsistent shadow boundary supervision and artefact generation, achieving enhanced accuracy and simplicity without requiring shadow masks during inference. Trained on a dataset of 10,000 natural images augmented with synthetic shadows, FieldNet outperforms state-of-the-art methods on benchmark datasets (ISTD, ISTD+, SRD), with up to $9$x speed improvements (66 FPS on Nvidia 2080Ti) and superior shadow removal quality (PSNR: 38.67, SSIM: 0.991). Real-world case studies in precision agriculture robotics demonstrate the practical impact of FieldNet in enhancing weed detection accuracy. These advancements establish FieldNet as a robust, efficient solution for real-time vision tasks in field robotics and beyond.

cs.CV↗

arXiv2024

Precision Robotic Spot-Spraying: Reducing Herbicide Use and Enhancing Environmental Outcomes in Sugarcane

Precise robotic weed control plays an essential role in precision agriculture. It can help significantly reduce the environmental impact of herbicides while reducing weed management costs for farmers. In this paper, we demonstrate that a custom-designed robotic spot spraying tool based on computer vision and deep learning can significantly reduce herbicide usage on sugarcane farms. We present results from field trials that compare robotic spot spraying against industry-standard broadcast spraying, by measuring the weed control efficacy, the reduction in herbicide usage, and the water quality improvements in irrigation runoff. The average results across 25 hectares of field trials show that spot spraying on sugarcane farms is 97\% as effective as broadcast spraying and reduces herbicide usage by 35\%, proportionally to the weed density. For specific trial strips with lower weed pressure, spot spraying reduced herbicide usage by up to 65\%. Water quality measurements of irrigation-induced runoff, three to six days after spraying, showed reductions in the mean concentration and mean load of herbicides of 39\% and 54\%, respectively, compared to broadcast spraying. These promising results reveal the capability of spot spraying technology to reduce herbicide usage on sugarcane farms without impacting weed control and potentially providing sustained water quality benefits.

cs.RO↗

arXiv2023

WeedCLR: Weed Contrastive Learning through Visual Representations with Class-Optimized Loss in Long-Tailed Datasets

Image classification is a crucial task in modern weed management and crop intervention technologies. However, the limited size, diversity, and balance of existing weed datasets hinder the development of deep learning models for generalizable weed identification. In addition, the expensive labelling requirements of mainstream fully-supervised weed classifiers make them cost- and time-prohibitive to deploy widely, for new weed species, and in site-specific weed management. This paper proposes a novel method for Weed Contrastive Learning through visual Representations (WeedCLR), that uses class-optimized loss with Von Neumann Entropy of deep representation for weed classification in long-tailed datasets. WeedCLR leverages self-supervised learning to learn rich and robust visual features without any labels and applies a class-optimized loss function to address the class imbalance problem in long-tailed datasets. WeedCLR is evaluated on two public weed datasets: CottonWeedID15, containing 15 weed species, and DeepWeeds, containing 8 weed species. WeedCLR achieves an average accuracy improvement of 4.3\% on CottonWeedID15 and 5.6\% on DeepWeeds over previous methods. It also demonstrates better generalization ability and robustness to different environmental conditions than existing methods without the need for expensive and time-consuming human annotations. These significant improvements make WeedCLR an effective tool for weed classification in long-tailed datasets and allows for more rapid and widespread deployment of site-specific weed management and crop intervention technologies.

cs.CV↗

arXiv2021

Analysis of a generalised Boltzmann equation for anomalous diffusion under time-dependent fields

The generalised Boltzmann equation which treats the combined localised and delocalised nature of transport present in certain materials is extended to accommodate time-dependent fields. In particular, AC fields are shown to be a means to probe the trapping and detrapping rates of materials under certain conditions. Conditions leading to dispersive transport are considered, and the signature of fractional/anomalous diffusion under AC electric fields is presented.

cond-mat.stat-mech↗

arXiv2019

A Survey of Asynchronous Programming Using Coroutines in the Internet of Things and Embedded Systems

Many Internet of Things and embedded projects are event-driven, and therefore require asynchronous and concurrent programming. Current proposals for C++20 suggest that coroutines will have native language support. It is timely to survey the current use of coroutines in embedded systems development. This paper investigates existing research which uses or describes coroutines on resource-constrained platforms. The existing research is analysed with regard to: software platform, hardware platform and capacity; use cases and intended benefits; and the application programming interface design used for coroutines. A systematic mapping study was performed, to select studies published between 2007 and 2018 which contained original research into the application of coroutines on resource-constrained platforms. An initial set of 566 candidate papers were reduced to only 35 after filters were applied, revealing the following taxonomy. The C & C++ programming languages were used by 22 studies out of 35. As regards hardware, 16 studies used 8- or 16-bit processors while 13 used 32-bit processors. The four most common use cases were concurrency (17 papers), network communication (15), sensor readings (9) and data flow (7). The leading intended benefits were code style and simplicity (12 papers), scheduling (9) and efficiency (8). A wide variety of techniques have been used to implement coroutines, including native macros, additional tool chain steps, new language features and non-portable assembly language. We conclude that there is widespread demand for coroutines on resource-constrained devices. Our findings suggest that there is significant demand for a formalised, stable, well-supported implementation of coroutines in C++, designed with consideration of the special needs of resource-constrained devices, and further that such an implementation would bring benefits specific to such devices.

cs.SE↗

arXiv2019

DeepWeeds: A Multiclass Weed Species Image Dataset for Deep Learning

Robotic weed control has seen increased research of late with its potential for boosting productivity in agriculture. Majority of works focus on developing robotics for croplands, ignoring the weed management problems facing rangeland stock farmers. Perhaps the greatest obstacle to widespread uptake of robotic weed control is the robust classification of weed species in their natural environment. The unparalleled successes of deep learning make it an ideal candidate for recognising various weed species in the complex rangeland environment. This work contributes the first large, public, multiclass image dataset of weed species from the Australian rangelands; allowing for the development of robust classification methods to make robotic weed control viable. The DeepWeeds dataset consists of 17,509 labelled images of eight nationally significant weed species native to eight locations across northern Australia. This paper presents a baseline for classification performance on the dataset using the benchmark deep learning models, Inception-v3 and ResNet-50. These models achieved an average classification accuracy of 95.1% and 95.7%, respectively. We also demonstrate real time performance of the ResNet-50 architecture, with an average inference time of 53.4 ms per image. These strong results bode well for future field implementation of robotic weed control methods in the Australian rangelands.

cs.CV↗

arXiv2017

Third-order transport coefficients for localised and delocalised charged-particle transport

We derive third order transport coefficients of skewness for a phase-space kinetic model that considers the processes of scattering collisions, trapping, detrapping and recombination losses. The resulting expression for the skewness tensor provides an extension to Fick's law which is in turn applied to yield a corresponding generalised advection-diffusion-skewness equation. A physical interpretation of trap-induced skewness is presented and used to describe an observed negative skewness due to traps. A relationship between skewness, diffusion, mobility and temperature is formed by analogy with Einstein's relation. Fractional transport is explored and its effects on the flux transport coefficients are also outlined.

cond-mat.stat-mech↗

arXiv2017

Generalised balance equations for charged particle transport via localised and delocalised states: Mobility, generalised Einstein relations and fractional transport

A generalised phase-space kinetic Boltzmann equation for highly non-equilibrium charged particle transport via localised and delocalised states is used to develop continuity, momentum and energy balance equations, accounting explicitly for scattering, trapping/detrapping and recombination loss processes. Analytic expressions detail the effect of these microscopic processes on the mobility and diffusivity. Generalised Einstein relations (GER) are developed that enable the anisotropic nature of diffusion to be determined in terms of the measured field-dependence of the mobility. Interesting phenomena such as negative differential conductivity and recombination heating/cooling are shown to arise from recombination loss processes and the localised and delocalised nature of transport. Fractional transport emerges naturally within this framework through the appropriate choice of divergent mean waiting time distributions for localised states, and fractional generalisations of the GER and mobility are presented. Signature impacts on time-of-flight current transients of recombination loss processes via both localised and delocalised states are presented.

cond-mat.stat-mech↗

arXiv2016

Solution of a generalised Boltzmann's equation for non-equilibrium charged particle transport via localised and delocalised states

We present a general phase-space kinetic model for charged particle transport through combined localised and delocalised states, capable of describing scattering collisions, trapping, detrapping and losses. The model is described by a generalised Boltzmann equation, for which an analytical solution is found in Fourier-Laplace space. The velocity of the centre of mass (CM) and the diffusivity about it are determined analytically, together with the flux transport coefficients. Transient negative values of the free particle CM transport coefficients can be observed due to the trapping to, and detrapping from, localised states. A Chapman-Enskog type perturbative solution technique is applied, confirming the analytical results and highlighting the emergence of a density gradient representation in the weak-gradient hydrodynamic regime. A generalised diffusion equation with a unique global time operator is shown to arise, reducing to the standard diffusion equation and a Caputo fractional diffusion equation in the normal and dispersive limits. A subordination transformation is used to solve the generalised diffusion equation by mapping from the solution of a corresponding standard diffusion equation.

cond-mat.stat-mech↗

arXiv2015

Efficient numerical solution of the time fractional diffusion equation by mapping from its Brownian counterpart

The solution of a Caputo time fractional diffusion equation of order $0<α<1$ is expressed in terms of the solution of a corresponding integer order diffusion equation. We demonstrate a linear time mapping between these solutions that allows for accelerated computation of the solution of the fractional order problem. In the context of an $N$-point finite difference time discretisation, the mapping allows for an improvement in time computational complexity from $O\left(N^2\right)$ to $O\left(N^α\right)$, given a precomputation of $O\left(N^{1+α}\ln N\right)$. The mapping is applied successfully to the least-squares fitting of a fractional advection diffusion model for the current in a time-of-flight experiment, resulting in a computational speed up in the range of one to three orders of magnitude for realistic problem sizes.

physics.comp-ph↗

arXiv2014

Molecular weight dependent bimolecular recombination in organic solar cells

Charge carrier recombination is studied in operational organic solar cells made from the polymer:fullerene system PCDTBT:PC71BM (poly[N-9"-heptadecanyl-2,7-carbazole-alt-5,5-(4',7'-di-2-thienyl-2',1',3'-benzothiadiazole)] : [6,6]-phenyl-C$_{70}$-butyric acid methyl ester). A newly developed technique High Intensity Resistance dependent PhotoVoltage (HI-RPV) is presented for reliably quantifying the bimolecular recombination coefficient independently of variations in experimental conditions, thereby resolving key limitations of previous experimental approaches. Experiments are performed on solar cells of varying thicknesses and varying polymeric molecular weights. It is shown that solar cells made from low molecular weight PCDTBT exhibit Langevin recombination, whereas suppressed (non-Langevin) recombination is found in solar cells made with high molecular weight PCDTBT.

cond-mat.mtrl-sci↗

arXiv2014

The impact of hot charge carrier mobility on photocurrent losses in polymer-based solar cells

A typical signature of charge extraction in disordered organic systems is dispersive transport, which implies a distribution of charge carrier mobilities that negatively impact on device performance. Dispersive transport has been commonly understood to originate from a time-dependent mobility of hot charge carriers that reduces as excess energy is lost during relaxation in the density of states. In contrast, we show via photon energy, electric field and film thickness independence of carrier mobilities that the dispersive photocurrent in organic solar cells originates not from the loss of excess energy during hot carrier thermalization, but rather from the loss of carrier density to trap states during transport. Our results emphasize that further efforts should be directed to minimizing the density of trap states, rather than controlling energetic relaxation of hot carriers within the density of states.

cond-mat.mtrl-sci↗

arXiv2011

Analytic solution of the fractional advection diffusion equation for the time-of-flight experiment in a finite geometry

A general analytic solution to the fractional advection diffusion equation is obtained in plane parallel geometry. The result is an infinite series of spatial Fourier modes which decay according to the Mittag-Leffler function, which is cast into a simple closed form expression in Laplace space using the Poisson summation theorem. An analytic expression for the current measured in a time-of-flight experiment is derived, and the sum of the slopes of the two respective time regimes on logarithmic axes is demonstrated to be -2, in agreement with the well known result for a continuous time random walk model. The sensitivity of current and particle number density to variation of experimentally controlled parameters is investigated in general, and the results applied to analyze selected experimental data.

cond-mat.stat-mech↗