Adam D. Kilpatrick
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Adam D. Kilpatrick has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Bertram Ostendorf
Department of Ecology and Evolutionary Biology, School of Biological Sciences, University of Adelaide, Adelaide, SA 5005, Australia
3 shared publications - John Llewellyn Read
School of Biological Sciences University of Adelaide Adelaide South Australia Australia
3 shared publications - Stephen Christopher Warren-Smith
Adelaide University
3 shared publications - Megan Lewis
School of Biological Sciences, The University of Adelaide, Adelaide, Australia
2 shared publications - Andrew Freeman1 shared publication
- David G. Lancaster
University of South Australia
1 shared publication
Latest Publications
- 13 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 42 (18) , 6342-6350
Dynamic Multimode Fiber Specklegram Sensor Smart Bed Enabled by Deep Learning
Fiber specklegram sensors eliminate the need for complex optical interrogators by using a camera to capture images or videos of the speckle pattern at the end facet of a multimode optical fiber. A fiber specklegram sensor is highly sensitive to movement anywhere along the length of the fiber, yielding complex information on external perturbations. The ability of the fiber specklegram sensor to be laid over a large surface area and respond to movement at any location can be useful in many real life sensing problems. Until now, only static speckle images have been used to evaluate such perturbations. However, there are applications where intrinsically dynamic measurands are desired, requiring different approaches to the analysis. In this work, we utilized deep learning techniques to analyze dynamic speckle videos for applications in dynamic biomechanical sensing, which we demonstrated for respiration rate monitoring as a proof-of-concept. This application demands both wide area sensing, that is, coverage of a mattress, together with dynamic information. We achieve this by recording videos of speckle from a single multimode fiber that covers a mattress in an S-configuration, and train convolutional neural networks directly on the video data. We show that the fiber specklegram sensor combined with our deep learning model can accurately classify the respiration rate. This approach has wide reaching potential for other biomechanical healthcare applications, such as pressure sore prevention and continuous monitoring to reduce the risk of falls.
- 13 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Journal of Lightwave Technology
- Vol. 42 (18) , 6342-6350
Dynamic Multimode Fiber Specklegram Sensor Smart Bed Enabled by Deep Learning
Fiber specklegram sensors eliminate the need for complex optical interrogators by using a camera to capture images or videos of the speckle pattern at the end facet of a multimode optical fiber. A fiber specklegram sensor is highly sensitive to movement anywhere along the length of the fiber, yielding complex information on external perturbations. The ability of the fiber specklegram sensor to be laid over a large surface area and respond to movement at any location can be useful in many real life sensing problems. Until now, only static speckle images have been used to evaluate such perturbations. However, there are applications where intrinsically dynamic measurands are desired, requiring different approaches to the analysis. In this work, we utilized deep learning techniques to analyze dynamic speckle videos for applications in dynamic biomechanical sensing, which we demonstrated for respiration rate monitoring as a proof-of-concept. This application demands both wide area sensing, that is, coverage of a mattress, together with dynamic information. We achieve this by recording videos of speckle from a single multimode fiber that covers a mattress in an S-configuration, and train convolutional neural networks directly on the video data. We show that the fiber specklegram sensor combined with our deep learning model can accurately classify the respiration rate. This approach has wide reaching potential for other biomechanical healthcare applications, such as pressure sore prevention and continuous monitoring to reduce the risk of falls.
- Open Access
- 3 February 2023
- Published by MDPI AG in ISPRS International Journal of Geo-Information
- Vol. 12 (2) , 50
A Spatial and Temporal Evaluation of Broad-Scale Yield Predictions Created from Yield Mapping Technology and Landsat Satellite Imagery in the Australian Mediterranean Dryland Cropping Region
Contemplation of potential strategies to adapt to a changing and variable climate in agricultural cropping areas depends on the availability of geo-information that is at a sufficient resolution, scale and temporal length to inform these decisions. We evaluated the efficacy of creating high-resolution, broad-scale indicators of yield from simple models that combine yield mapping data, a precision agriculture tool, with the normalised difference vegetation index (NDVI) from Landsat 5 and 7 ETM+ imagery. These models were then generalised to test its potential operationalisation across a large agricultural region (>1/2 million hectares) and the state of South Australia (>8 million hectares). Annual models were the best predictors of yield across both areas. Moderate discrimination accuracy in the regional analysis meant that models could be extrapolated with reasonable spatial precision, whereas the accuracy across the state-wide analysis was poor. Generalisation of these models to further operationalise the methodology by removing the need for crop type discrimination and the continual access to annual yield data showed some benefit. The application of this approach with past and contemporary datasets can create a long-term archive that fills an information void, providing a powerful evidence base to inform current management decisions and future on-farm land use in cropping regions elsewhere.
- 3 February 2023
- Published by MDPI AG in ISPRS International Journal of Geo-Information
- Vol. 12 (2) , 50
Open AccessA Spatial and Temporal Evaluation of Broad-Scale Yield Predictions Created from Yield Mapping Technology and Landsat Satellite Imagery in the Australian Mediterranean Dryland Cropping Region
Contemplation of potential strategies to adapt to a changing and variable climate in agricultural cropping areas depends on the availability of geo-information that is at a sufficient resolution, scale and temporal length to inform these decisions. We evaluated the efficacy of creating high-resolution, broad-scale indicators of yield from simple models that combine yield mapping data, a precision agriculture tool, with the normalised difference vegetation index (NDVI) from Landsat 5 and 7 ETM+ imagery. These models were then generalised to test its potential operationalisation across a large agricultural region (>1/2 million hectares) and the state of South Australia (>8 million hectares). Annual models were the best predictors of yield across both areas. Moderate discrimination accuracy in the regional analysis meant that models could be extrapolated with reasonable spatial precision, whereas the accuracy across the state-wide analysis was poor. Generalisation of these models to further operationalise the methodology by removing the need for crop type discrimination and the continual access to annual yield data showed some benefit. The application of this approach with past and contemporary datasets can create a long-term archive that fills an information void, providing a powerful evidence base to inform current management decisions and future on-farm land use in cropping regions elsewhere.
- Open Access
- 24 June 2022
- Published by SPIE-Intl Soc Optical Eng in Journal of Biomedical Optics
- Vol. 27 (06) , 067002-12
Multimode optical fiber specklegram smart bed sensor array
Significance: Monitoring the movement and vital signs of patients in hospitals and other healthcare environments is a significant burden on healthcare staff. Early warning systems using smart bed sensors hold promise to relieve this burden and improve patient outcomes. We propose a scalable and cost-effective optical fiber sensor array that can be embedded into a mattress to detect movement, both sensitively and spatially. Aim: Proof-of-concept demonstration that a multimode optical fiber (MMF) specklegram sensor array can be used to detect and image movement on a bed. Approach: Seven MMFs are attached to the upper surface of a mattress such that they cross in a 3 × 4 array. The specklegram output is monitored using a single laser and single camera and movement on the fibers is monitored by calculating a rolling zero-normalized cross-correlation. A 3 × 4 image is formed by comparing the signal at each crossing point between two fibers. Results: The MMF sensor array can detect and image movement on a bed, including getting on and off the bed, rolling on the bed, and breathing. Conclusions: The sensor array shows a high sensitivity to movement, which can be used for monitoring physiological parameters and patient movement for potential applications in healthcare settings.
- 24 June 2022
- Published by SPIE-Intl Soc Optical Eng in Journal of Biomedical Optics
- Vol. 27 (06) , 067002-12
Open AccessMultimode optical fiber specklegram smart bed sensor array
Significance: Monitoring the movement and vital signs of patients in hospitals and other healthcare environments is a significant burden on healthcare staff. Early warning systems using smart bed sensors hold promise to relieve this burden and improve patient outcomes. We propose a scalable and cost-effective optical fiber sensor array that can be embedded into a mattress to detect movement, both sensitively and spatially. Aim: Proof-of-concept demonstration that a multimode optical fiber (MMF) specklegram sensor array can be used to detect and image movement on a bed. Approach: Seven MMFs are attached to the upper surface of a mattress such that they cross in a 3 × 4 array. The specklegram output is monitored using a single laser and single camera and movement on the fibers is monitored by calculating a rolling zero-normalized cross-correlation. A 3 × 4 image is formed by comparing the signal at each crossing point between two fibers. Results: The MMF sensor array can detect and image movement on a bed, including getting on and off the bed, rolling on the bed, and breathing. Conclusions: The sensor array shows a high sensitivity to movement, which can be used for monitoring physiological parameters and patient movement for potential applications in healthcare settings.
- Open Access
- 13 November 2015
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 10 (11) , e0142742
Rangeland Condition Monitoring: A New Approach Using Cross-Fence Comparisons of Remotely Sensed Vegetation
A need exists in arid rangelands for effective monitoring of the impacts of grazing management on vegetation cover. Monitoring methods which utilize remotely-sensed imagery may have comprehensive spatial and temporal sampling, but do not necessarily control for spatial variation of natural variables, such as landsystem, vegetation type, soil type and rainfall. We use the inverse of the red band from Landsat TM satellite imagery to determine levels of vegetation cover in a 22,672km2 area of arid rangeland in central South Australia. We interpret this wealth of data using a cross-fence comparison methodology, allowing us to rank paddocks (fields) in the study region according to effectiveness of grazing management. The cross-fence comparison methodology generates and solves simultaneous equations of the relationship between each paddock and all other paddocks, derived from pairs of cross-fence sample points. We compare this ranking from two image dates separated by six years, during which management changes are known to have taken place. Changes in paddock rank resulting from the cross-fence comparison method show strong correspondence to those predicted by grazing management in this region, with a significant difference between the two common management types; a change from full stocking rate to light 20% stocking regime (Major Stocking Reduction) and maintenance of full 100% stocking regime (Full Stocking Maintained) (P = 0.00000132). While no paddocks had a known increase in stocking rate during the study period, many had a reduction or complete removal in stock numbers, and many also experienced removals of pest species, such as rabbits, and other ecosystem restoration activities. These paddocks generally showed an improvement in rank compared to paddocks where the stocking regime remained relatively unchanged. For the first time, this method allows us to rank non-adjacent paddocks in a rangeland region relative to each other, while controlling for natural spatio-temporal variables such as rainfall, soil type, and vegetation community distributions, due to the nature of the cross-fence experimental design, and the spatially comprehensive data available in satellite imagery. This method provides a potential tool to aid land managers in decision making processes, particularly with regard to stocking rates.
- 13 November 2015
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 10 (11) , e0142742
Open AccessRangeland Condition Monitoring: A New Approach Using Cross-Fence Comparisons of Remotely Sensed Vegetation
A need exists in arid rangelands for effective monitoring of the impacts of grazing management on vegetation cover. Monitoring methods which utilize remotely-sensed imagery may have comprehensive spatial and temporal sampling, but do not necessarily control for spatial variation of natural variables, such as landsystem, vegetation type, soil type and rainfall. We use the inverse of the red band from Landsat TM satellite imagery to determine levels of vegetation cover in a 22,672km2 area of arid rangeland in central South Australia. We interpret this wealth of data using a cross-fence comparison methodology, allowing us to rank paddocks (fields) in the study region according to effectiveness of grazing management. The cross-fence comparison methodology generates and solves simultaneous equations of the relationship between each paddock and all other paddocks, derived from pairs of cross-fence sample points. We compare this ranking from two image dates separated by six years, during which management changes are known to have taken place. Changes in paddock rank resulting from the cross-fence comparison method show strong correspondence to those predicted by grazing management in this region, with a significant difference between the two common management types; a change from full stocking rate to light 20% stocking regime (Major Stocking Reduction) and maintenance of full 100% stocking regime (Full Stocking Maintained) (P = 0.00000132). While no paddocks had a known increase in stocking rate during the study period, many had a reduction or complete removal in stock numbers, and many also experienced removals of pest species, such as rabbits, and other ecosystem restoration activities. These paddocks generally showed an improvement in rank compared to paddocks where the stocking regime remained relatively unchanged. For the first time, this method allows us to rank non-adjacent paddocks in a rangeland region relative to each other, while controlling for natural spatio-temporal variables such as rainfall, soil type, and vegetation community distributions, due to the nature of the cross-fence experimental design, and the spatially comprehensive data available in satellite imagery. This method provides a potential tool to aid land managers in decision making processes, particularly with regard to stocking rates.
- 31 March 2011
- Published by Elsevier in Ecological Informatics
- Vol. 6 (2) , 170-176
Cross-fence comparisons: Theory for spatially comprehensive, controlled variable assessment of treatment effects in managed landscapes
- 31 March 2011
- Published by Elsevier in Ecological Informatics
- Vol. 6 (2) , 170-176
Cross-fence comparisons: Theory for spatially comprehensive, controlled variable assessment of treatment effects in managed landscapes
- 28 March 2011
- Published by Wiley in Ecological Management & Restoration
- Vol. 12 (1) , 46-53
Eradication of rabbits from landscape scale exclosures: pipedream or possibility?
- 28 March 2011
- Published by Wiley in Ecological Management & Restoration
- Vol. 12 (1) , 46-53
Eradication of rabbits from landscape scale exclosures: pipedream or possibility?
- 1 January 2009
- Published by Taylor & Francis in Transactions of the Royal Society of South Australia
- Vol. 133 (1) , 172-177
Improved Conservation Status ofAtriplex KochianaMaiden
Five new populations and a significant expansion of the Andamooka opal fields population of Atriplex kochiana have considerably increased the range and known population of this rare saltbush. Evidence from repeated monitoring of these populations has confirmed that A. kochiana is a short-lived saltbush that can regenerate from seed following summer rains. No evidence that either cattle or rabbits had impacted the monitored individuals was detected and regenerating populations on track verges and erosion gutters suggest the species is relatively robust to disturbance.
- 1 January 2009
- Published by Taylor & Francis in Transactions of the Royal Society of South Australia
- Vol. 133 (1) , 172-177
Improved Conservation Status ofAtriplex KochianaMaiden
Five new populations and a significant expansion of the Andamooka opal fields population of Atriplex kochiana have considerably increased the range and known population of this rare saltbush. Evidence from repeated monitoring of these populations has confirmed that A. kochiana is a short-lived saltbush that can regenerate from seed following summer rains. No evidence that either cattle or rabbits had impacted the monitored individuals was detected and regenerating populations on track verges and erosion gutters suggest the species is relatively robust to disturbance.