arXiv subjects

Amanda Ramcharan

Publications and source records attributed to Amanda Ramcharan.

3 recordsLinked to original sources

arXiv2018

Assessing a mobile-based deep learning model for plant disease surveillance

Convolutional neural network models (CNNs) have made major advances in computer vision tasks in the last five years. Given the challenge in collecting real world datasets, most studies report performance metrics based on available research datasets. In scenarios where CNNs are to be deployed on images or videos from mobile devices, models are presented with new challenges due to lighting, angle, and camera specifications, which are not accounted for in research datasets. It is essential for assessment to also be conducted on real world datasets if such models are to be reliably integrated with products and services in society. Plant disease datasets can be used to test CNNs in real time and gain insight into real world performance. We train a CNN object detection model to identify foliar symptoms of diseases (or lack thereof) in cassava (Manihot esculenta Crantz). We then deploy the model on a mobile app and test its performance on mobile images and video of 720 diseased leaflets in an agricultural field in Tanzania. Within each disease category we test two levels of severity of symptoms - mild and pronounced, to assess the model performance for early detection of symptoms. In both severities we see a decrease in the F-1 score for real world images and video. The F-1 score dropped by 32% for pronounced symptoms in real world images (the closest data to the training data) due to a drop in model recall. If the potential of smartphone CNNs are to be realized our data suggest it is crucial to consider tuning precision and recall performance in order to achieve the desired performance in real world settings. In addition, the varied performance related to different input data (image or video) is an important consideration for the design of CNNs in real world applications.

cs.CV↗

arXiv2018

Soil Property and Class Maps of the Conterminous US at 100 meter Spatial Resolution based on a Compilation of National Soil Point Observations and Machine Learning

With growing concern for the depletion of soil resources, conventional soil data must be updated to support spatially explicit human-landscape models. Three US soil point datasetswere combined with a stack of over 200 environmental datasets to generate complete coverage gridded predictions at 100 m spatial resolution of soil properties (percent organic C, total N, bulk density, pH, and percent sand and clay) and US soil taxonomic classes (291 great groups and 78 modified particle size classes) for the conterminous US. Models were built using parallelized random forest and gradient boosting algorithms. Soil property predictions were generated at seven standard soil depths (0, 5, 15, 30, 60, 100 and 200 cm). Prediction probability maps for US soil taxonomic classifications were also generated. Model validation results indicate an out-of-bag classification accuracy of 60 percent for great groups, and 66 percent for modified particle size classes; for soil properties cross-validated R-square ranged from 62 percent for total N to 87 percent for pH. Nine independent validation datasets were used to assess prediction accuracies for soil class models and results ranged between 24-58 percent and 24-93 percent for great group and modified particle size class prediction accuracies, respectively. The hybrid "SoilGrids+" modeling system that incorporates remote sensing data, local predictions of soil properties, conventional soil polygon maps, and machine learning opens the possibility for updating conventional soil survey data with machine learning technology to make soil information easier to integrate with spatially explicit models, compared to multi-component map units.

physics.geo-ph↗

arXiv2017

Using Transfer Learning for Image-Based Cassava Disease Detection

Cassava is the third largest source of carbohydrates for human food in the world but is vulnerable to virus diseases, which threaten to destabilize food security in sub-Saharan Africa. Novel methods of cassava disease detection are needed to support improved control which will prevent this crisis. Image recognition offers both a cost effective and scalable technology for disease detection. New transfer learning methods offer an avenue for this technology to be easily deployed on mobile devices. Using a dataset of cassava disease images taken in the field in Tanzania, we applied transfer learning to train a deep convolutional neural network to identify three diseases and two types of pest damage (or lack thereof). The best trained model accuracies were 98% for brown leaf spot (BLS), 96% for red mite damage (RMD), 95% for green mite damage (GMD), 98% for cassava brown streak disease (CBSD), and 96% for cassava mosaic disease (CMD). The best model achieved an overall accuracy of 93% for data not used in the training process. Our results show that the transfer learning approach for image recognition of field images offers a fast, affordable, and easily deployable strategy for digital plant disease detection.

cs.CV↗