https://hdl.handle.net/10680/1603 2023-03-05T02:51:34Z 2023-03-05T02:51:34Z Mostafa, Sakib Mondal, Debajyoti Beck, Michael A. Bidinosti, Christopher P. Henry, Christopher J. Stavness, Ian https://hdl.handle.net/10680/2000 2022-06-04T07:01:15Z 2022 - 05 - 11 - t00:00:00z

Leveraging Guided Backpropagation to Select Convolutional Neural Networks for Plant Classification Mostafa, Sakib; Mondal, Debajyoti; Beck, Michael A.; Bidinosti, Christopher P.; Henry, Christopher J.; Stavness, Ian The development of state-of-the-art convolutional neural networks (CNN) has allowed researchers to perform plant classification tasks previously thought impossible and rely on human judgment. Researchers often develop complex CNN models to achieve better performances, introducing over-parameterization and forcing the model to overfit on a training dataset. The most popular process for evaluating overfitting in a deep learning model is using accuracy and loss curves. Train and loss curves may help understand the performance of a model but do not provide guidance on how the model could be modified to attain better performance. In this article, we analyzed the relation between the features learned by a model and its capacity and showed that a model with higher representational capacity might learn many subtle features that may negatively affect its performance. Next, we showed that the shallow layers of a deep learning model learn more diverse features than the ones learned by the deeper layers. Finally, we propose SSIM cut curve, a new way to select the depth of a CNN model by using the pairwise similarity matrix between the visualization of the features learned at different depths by using Guided Backpropagation. We showed that our proposed method could potentially pave a new way to select a better CNN model.

2022 - 05 - 11 - t00:00:00z Yu, Jiajie Henry, Christopher J. https://hdl.handle.net/10680/1973 2021-10-26T07:01:14Z 2019 - 02年- 02 - t00:00:00z

Descriptive Topological Spaces for Performing Visual Search Yu, Jiajie; Henry, Christopher J. This article presents an approach to performing the task of visual search in the context of descriptive topological spaces. The presented algorithm forms the basis of a descriptive visual search system (DVSS) that is based on the guided search model (GSM) that is motivated by human visual search. This model, in turn, consists of the bottom-up and top-down attention models and is implemented within the DVSS in three distinct stages. First, the bottom-up activation process is used to generate saliency maps and to identify salient objects. Second, perceptual objects, defined in the context of descriptive topological spaces, are identified and associated with feature vectors obtained from a VGG deep learning convolutional neural network. Lastly, the top-down activation process makes decisions on whether the object of interest is present in a given image through the use of descriptive patterns within the context of a descriptive topological space. The presented approach is tested with images from the ImageNet ILSVRC2012 and SIMPLIcity datasets. The contribution of this article is a descriptive pattern-based visual search algorithm. Accepted version

2019 - 02年- 02 - t00:00:00z Henry, Christopher J. Awais, Syed Aqeel https://hdl.handle.net/10680/1972 2021-10-19T07:00:45Z 2018-11-03T00:00:00Z

A Descriptive Tolerance Nearness Measure for Performing Graph Comparison Henry, Christopher J.; Awais, Syed Aqeel This article proposes the tolerance nearness measure (TNM) as a computationally reduced alternative to the graph edit distance (GED) for performing graph comparisons. The TNM is defined within the context of near set theory, where the central idea is that determining similarity between sets of disjoint objects is at once intuitive and practically applicable. The TNM between two graphs is produced using the Bron-Kerbosh maximal clique enumeration algorithm. The result is that the TNM approach is less computationally complex than the bipartite-based GED algorithm. The contribution of this paper is the application of TNM to the problem of quantifying the similarity of disjoint graphs and that the maximal clique enumeration-based TNM produces comparable results to the GED when applied to the problem of content-based image processing, which becomes important as the number of nodes in a graph increases. Accepted version

2018-11-03T00:00:00Z Beck, Michael A. Liu, Chen-Yi Bidinosti, Christopher P. Henry, Christopher J. Godee, Cara M. Ajmani, Manisha https://hdl.handle.net/10680/1882 2021-05-12T19:45:40Z 2020-12-17T00:00:00Z

嵌入式系统的自动生成labeled plant images to enable machine learning applications in agriculture Beck, Michael A.; Liu, Chen-Yi; Bidinosti, Christopher P.; Henry, Christopher J.; Godee, Cara M.; Ajmani, Manisha A lack of sufficient training data, both in terms of variety and quantity, is often the bottleneck in the development of machine learning (ML) applications in any domain. For agricultural applications, ML-based models designed to perform tasks such as autonomous plant classification will typically be coupled to just one or perhaps a few plant species. As a consequence, each crop-specific task is very likely to require its own specialized training data, and the question of how to serve this need for data now often overshadows the more routine exercise of actually training such models. To tackle this problem, we have developed an embedded robotic system to automatically generate and label large datasets of plant images for ML applications in agriculture. The system can image plants from virtually any angle, thereby ensuring a wide variety of data; and with an imaging rate of up to one image per second, it can produce lableled datasets on the scale of thousands to tens of thousands of images per day. As such, this system offers an important alternative to time- and costintensive methods of manual generation and labeling. Furthermore, the use of a uniform background made of blue keying fabric enables additional image processing techniques such as background replacement and image segementation. It also helps in the training process, essentially forcing the model to focus on the plant features and eliminating random correlations. To demonstrate the capabilities of our system, we generated a dataset of over 34,000 labeled images, with which we trained an ML-model to distinguish grasses from nongrasses in test data from a variety of sources. We now plan to generate much larger datasets of Canadian crop plants and weeds that will be made publicly available in the hope of further enabling ML applications in the agriculture sector.

2020-12-17T00:00:00Z