Agriculture and Food Systems - Research Publications
Permanent URI for this collection
Search Results
1 - 10 of 1468
-
ItemMethodologies Used in Remote Sensing Data Analysis and Remote Sensors for Precision Agriculture(MDPI, 2022-10-01)When adopting remote sensing techniques in precision agriculture, there are two main areas to consider: data acquisition and data analysis methodologies [...].
-
ItemEarly Detection of Fusarium oxysporum Infection in Processing Tomatoes (Solanum lycopersicum) and Pathogen-Soil Interactions Using a Low-Cost Portable Electronic Nose and Machine Learning Modeling(MDPI, 2022-11-01)The early detection of pathogen infections in plants has become an important aspect of integrated disease management. Although previous research demonstrated the idea of applying digital technologies to monitor and predict plant health status, there is no effective system for detecting pathogen infection before symptomatology appears. This paper presents the use of a low-cost and portable electronic nose coupled with machine learning (ML) models for early disease detection. Several artificial neural network models were developed to predict plant physiological data and classify processing tomato plants and soil samples according to different levels of pathogen inoculum by using e-nose outputs as inputs, plant physiological data, and the level of infection as targets. Results showed that the pattern recognition models based on different infection levels had an overall accuracy of 94.4-96.8% for tomato plants and between 94.81% and 96.22% for soil samples. For the prediction of plant physiological parameters (photosynthesis, stomatal conductance, and transpiration) using regression models or tomato plants, the overall correlation coefficient was 0.97-0.99, with very significant slope values in the range 0.97-1. The performance of all models shows no signs of under or overfitting. It is hence proven accurate and valid to use the electronic nose coupled with ML modeling for effective early disease detection of processing tomatoes and could also be further implemented to monitor other abiotic and biotic stressors.
-
ItemLivestock Identification Using Deep Learning for Traceability(MDPI, 2022-11-01)Farm livestock identification and welfare assessment using non-invasive digital technology have gained interest in agriculture in the last decade, especially for accurate traceability. This study aimed to develop a face recognition system for dairy farm cows using advanced deep-learning models and computer vision techniques. This approach is non-invasive and potentially applicable to other farm animals of importance for identification and welfare assessment. The video analysis pipeline follows standard human face recognition systems made of four significant steps: (i) face detection, (ii) face cropping, (iii) face encoding, and (iv) face lookup. Three deep learning (DL) models were used within the analysis pipeline: (i) face detector, (ii) landmark predictor, and (iii) face encoder. All DL models were finetuned through transfer learning on a dairy cow dataset collected from a robotic dairy farm located in the Dookie campus at The University of Melbourne, Australia. Results showed that the accuracy across videos from 89 different dairy cows achieved an overall accuracy of 84%. The computer program developed may be deployed on edge devices, and it was tested on NVIDIA Jetson Nano board with a camera stream. Furthermore, it could be integrated into welfare assessment previously developed by our research group.
-
ItemRapid Detection of Fraudulent Rice Using Low-Cost Digital Sensing Devices and Machine Learning(MDPI, 2022-11-01)Rice fraud is one of the common threats to the rice industry. Conventional methods to detect rice adulteration are costly, time-consuming, and tedious. This study proposes the quantitative prediction of rice adulteration levels measured through the packaging using a handheld near-infrared (NIR) spectrometer and electronic nose (e-nose) sensors measuring directly on samples and paired with machine learning (ML) algorithms. For these purposes, the samples were prepared by mixing rice at different ratios from 0% to 100% with a 10% increment based on the rice's weight, consisting of (i) rice from different origins, (ii) premium with regular rice, (iii) aromatic with non-aromatic, and (iv) organic with non-organic rice. Multivariate data analysis was used to explore the sample distribution and its relationship with the e-nose sensors for parameter engineering before ML modeling. Artificial neural network (ANN) algorithms were used to predict the adulteration levels of the rice samples using the e-nose sensors and NIR absorbances readings as inputs. Results showed that both sensing devices could detect rice adulteration at different mixing ratios with high correlation coefficients through direct (e-nose; R = 0.94-0.98) and non-invasive measurement through the packaging (NIR; R = 0.95-0.98). The proposed method uses low-cost, rapid, and portable sensing devices coupled with ML that have shown to be reliable and accurate to increase the efficiency of rice fraud detection through the rice production chain.
-
ItemInfrared Thermal Imaging and Morpho-Physiological Indices Used for Wheat Genotypes Screening under Drought and Heat Stress(MDPI, 2022-12-01)Bread wheat, one of the largest broadacre crops, often experiences various environmental stresses during critical growth stages. Terminal drought and heat stress are the primary causes of wheat yield reduction worldwide. This study aimed to determine the drought and heat stress tolerance level of a group of 46 diverse wheat genotypes procured from the Australian Grains Gene Bank, Horsham, VIC Australia. Two separate drought stress (DS) and heat stress (HS) pot experiments were conducted in separate growth chambers. Ten days after complete anthesis, drought (40 ± 3% field capacity for 14 days) and heat stress (36/22 °C for three consecutive days) were induced. A significant genotype × environment interaction was observed and explained by various morpho-physiological traits, including rapid, non-destructive infrared thermal imaging for computational water stress indices. Except for a spike length in DS and harvest index in HS, the analysis of variance showed significant differences for all the recorded traits. Results showed grains per spike, grains weight per spike, spike fertility, delayed flag leaf senescence, and cooler canopy temperature were positively associated with grain yield under DS and HS. The flag leaf senescence and chlorophyll fluorescence were used to measure each genotype's stay-green phenotype and photosystem II activity after DS and HS. This study identified the top ten best and five lowest-performing genotypes from drought and heat stress experiments based on their overall performance. Results suggest that if heat or drought adaptive traits are brought together in a single genotype, grain yield can be improved further, particularly in a rainfed cropping environment.
-
ItemBenefits of prolonged ageing for the quality of Australian pork depends on cooking temperature and meat pH(CSIRO Publishing, 2023)Context. Heating of meat leads to structural changes reflected in the juiciness and the tenderness of the cooked meat. Aims. This study aimed to characterise the effect of prolonged ageing and cooking on pork-quality traits. Methods. Longissimus lumborum samples from 12 carcasses were aged 3 days (conventional ageing) or 15 days (prolonged ageing) and pork cuboids were cooked at 50–80°C for 30 min. Cooking loss, total water content (TWC), Warner–Bratzler shear force (WBSF) and shrinkage (longitudinal, transverse and estimated volume) of the pork loin cuboids were measured. Key results. Prolonged ageing for 15 days reduced the WBSF of samples cooked at 50°C, and the cooking loss for samples cooked at 70°Cand 80°C, relative to conventional ageing for 3 days. The WBSF of pork aged for 15 days was not different from that of pork aged for 3 days. Prolonged ageing reduced longitudinal shrinkage of cuboids, but TWC and transverse/ volume shrinkage of cuboids were not affected by ageing. The diameter of cooked muscle fibre fragments was smaller in pork subjected to prolonged ageing. Conclusions. Prolonged ageing was favourable for minimising cooking loss at higher cooking temperatures but was only favourable for tenderness at the lowest cooking temperature . Low pH of the samples is likely to have caused the lack of tenderisation with ageing.
-
ItemIncreased light scattering in electrically stimulated beef longissimus muscle fibres contributes to the observed meat colour at grading(CSIRO PUBLISHING, 2023-02-20)Context: Electrical stimulation is often used by meat processors to promote fast muscle pH decline and optimise meat quality. Meat colour can be made more acceptable by this process, but how this relates to the microstructure and light-scattering properties of muscle is still unknown. Aims: To investigate the effect of electrical stimulation of beef carcasses on the meat colour at grading and the role of the muscle fibre microstructure and light scattering in determining colour differences. Methods: Electrical stimulation inputs (electrical stimulation inputs (ES), n = 8; no electrical stimulation inputs (NS), n = 8) were applied to beef carcasses from female cattle of approximately 18–24 months of age. ES comprised electrical immobilisation, bleed rail electric simulation and hide puller rigidity probe, which have all been shown to increase pH fall post-mortem in beef carcasses. pH fall was monitored, the longissimus thoracis was graded at 20–22 h postmortem and measurements were made of colour, muscle-fibre structure and light scattering. Key results: The decline of pH was increased in ES relative to NS, as indicated by lower pH at 2 h postmortem (5.83 vs 6.86 respectively, s.e. = 0.068; P < 0.05) as well as changes in both chromatic colour a* b* and achromatic (no colour) lightness in the muscle. Chromatic changes were evident as higher grader colour scores, increased redness (a*) and yellowness (b*) with higher levels of oxymyoglobin and lower levels of deoxymyoglobin. Achromatic changes were evident as increased lightness (L*) and surface reflectance (%R) at the meat surface, and increased global brightness within the muscle fibres. Conclusions: Increased lightness and brightness in electrically stimulated muscles were likely to be due to formation of contraction nodes and distortion of muscle fibres, which changed the microstructure of muscle in ways that increased its light-scattering properties. Implications: Consideration of the role of light scattering in determining beef colour at grading will advance understanding of how to improve this important quality trait.
-
ItemSelecting methods of agricultural extension to support diverse adoption pathways: a review and case studies(CSIRO PUBLISHING, 2022-12-23)This paper presents results from a review of methods of agricultural extension, including the evidence for the effectiveness of methods in supporting farm practice change, how they affect the change process, and the critical success factors involved. Agricultural scientists face challenges in aligning their research outputs to the change process on farm. These challenges are exacerbated by the funding environment for research, development, and extension (RD&E), the complexity of the adoption process and the privatisation and commercialisation of advisory and extension services. To assist scientists in navigating these challenges, a structured literature review of extension methods was conducted, examining the following: group-learning/peer-topeer; technology development; training; information provision; one-on-one advice/coaching; e-extension; co-innovation; best management practice; and social marketing. In addition, two case studies outlining the application of combinations of extension methods in the context of feeding system challenges in the Australian dairy industry, and their effects, are described. While the evidence across the studies reviewed was strongest for the effect on adoption of small group-learning and one to one consulting, it was combinations of methods that resulted in larger effects (for example, in practice change or profitability), which was credited to how they addressed the human and social dimensions of the adoption process. Case studies of adoption in the dairy sector found that scientists influenced adoption by collaborating with the private sector, being directly involved with on-farm trials and demonstrations, and supporting grouplearning approaches to help the adoption of past research. This role for scientists in adoption was enabled by investment in programs of RD&E rather than discreet research experiments, and research designs and methods that incorporated the social dimensions of adoption. This synthesis demonstrates the need for scientists to be proactive in providing guidance for farmers on where to access and source information related to their work, engage with a broad range of advisor types associated with their research field, champion in-field trials and/or demonstrations and be active participants in collaborative approaches to RD&E.
-
ItemIntegrating nutrition and obesity prevention considerations into institutional investment decisions regarding food companies: Australian investment sector perspectives(BMC, 2022-11-08)BACKGROUND: There is growing recognition that current food systems are both unhealthy and unsustainable, and are increasingly shifting toward the supply and marketing of unhealthy, ultra-processed foods and beverages. Large food companies hold substantial power within food systems and present a significant barrier to progress on addressing issues related to nutrition and obesity prevention. Institutional investors (such as pension funds) play a key role in influencing corporate governance and practices, and are increasingly incorporating environmental, social and governance (ESG) considerations within investment decisions. By considering nutrition and obesity prevention, institutional investors present a potential avenue for driving increased food industry accountability for their population health impact. This study investigated views of stakeholders in the Australian investment sector on the incorporation of nutrition and obesity prevention considerations within institutional investment decision-making regarding food companies. METHODS: Fifteen in-depth, semi-structured interviews were conducted in 2020-21. Participants were predominantly Australian-based, and included representatives from asset management companies, superannuation funds, ESG advisory/consultancy firms, ESG research providers, and relevant advocacy groups. Interviews examined challenges and opportunities to the integration of nutrition and obesity prevention considerations within institutional investment decision-making. Interviews were analysed using deductive thematic analysis, informed by a theoretical change model. RESULTS: Several participants reported that their institution factored nutrition and obesity prevention considerations into their investment decisions; however, attention to nutrition-related issues was limited, generally perceived as 'niche', and not yet institutionalised. Key challenges and opportunities were identified at the employee, investment organisation, investment sector, government and non-government levels. These challenges and opportunities centred around experience and knowledge, quality and availability of ESG data and benchmarks, importance of investor coalitions, and demonstration of financial risks related to nutrition and obesity. CONCLUSION: There are a range of steps that could be taken to help ensure more systematic and effective consideration of issues related to nutrition and obesity prevention within institutional investment decision-making in Australia, including: (1) improved nutrition-related reporting metrics and benchmarking criteria for food companies; (2) better articulation of the financial risks that unhealthy diets and obesity pose to investors; (3) enhanced investor advocacy on unhealthy diets and obesity through investor coalitions and; (4) detailed guidance for investors on how to address unhealthy diets and obesity. Better engagement between the Australian public health community, institutional investors and government regulators is critical to drive changed investor practice in this area.
-
ItemNo Preview AvailableCharacterization of Fe(III)-binding peptides from pea protein hydrolysates targeting enhanced iron bioavailability(Elsevier BV, 2023-03)This investigation aimed to characterize enzymatically-derived pea peptides that act as solubilizing agents to enhance Fe(III) bioaccessibility. The pea hydrolysates were found to have an iron-binding capacity of 5.3 mg/g lyophilized powder. The Fe(III)-binding peptides were separated by immobilized metal affinity chromatography (IMAC), and then sequenced using tandem MS following an in-solution tryptic digestion. Results revealed that the Fe(III)-binding fraction was rich in Glu, Asn, Lys and Leu, and the peptides primarily belonged to the Vicilin family. After screening based on the peptides’ relative abundance and physicochemical properties, 15 novel peptides below 1.5 kDa were identified as potential candidates for enhancement of iron bioavailability. Fourier Transform infrared spectroscopy (FTIR) of the hydrolysate-iron complex suggested that the principal sites of peptide binding corresponded primarily to the carboxylate groups, with amine I and II groups also evident.