Browsing by Subject "Visual attention"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Automated Wheat Stem Rust Detection using Computer Vision
    (2023-05) Mahesh, Rahul Moorthy
    Wheat is one of the most important cereal crops, contributing significantly to the financial economy and food sources. Currently, the direct consumption of wheat amounts to about 41%. Additionally, in 2019 alone, the global trade value of wheat was about $39.6 billion. Hence, the protection of the yield of such crops from diseases is of immense importance. Stem rust is a fungal disease that attacks cereal crops. In particular, it is a common disease that occurs in wheat and destroys 50 to 70% of the yield if left unchecked. The loss of yield would in turn affect the economy and food consumption. Thus, there is a need to detect the outbreak early to apply fungicide treatment to the field. The traditional approach for detection involves experts inspecting the fields visually and grading them for stem rust which is a time-consuming process for a large field and can also be affected by human errors. Hence, an automated approach to the grading process would help solve such problems. The availability of an automated grading process will allow mobile robots, popularly being used for activities like irrigation, seed sowing, and precision agriculture to rapidly perform grading and alert the experts in case of detected stem rust. The alert through the automated detection would in turn lead to a timely application of fungicide for preventing the spread of stem rust in an efficient manner. The thesis focuses on formulating the wheat rust grading as a multi-class classification problem and demonstrating the effectiveness of the visual attention approach for solving it. The thesis also presents the first RGB field dataset with labels from experts for the development of automated stem rust grading approaches. The proposed approach was developed and evaluated on the presented dataset and shows the ability to distinguish between different intensities of stem rust with 86% accuracy. The reliability of the network is also validated qualitatively through attention maps where the visual attention approach shows interpretable focus areas compared to traditional detection approaches which fail to identify the general presence area of stem rust.
  • Thumbnail Image
    Item
    Emergence of human cognition from spatial and temporal structures
    (2012-09) Lin, Zhicheng
    Like solving a jigsaw puzzle by fitting pieces together bit by bit, the human mind makes sense of the world by transforming inputs into outputs throughout the nervous system, with the emergence of visual descriptions during such transformations relying heavily on perceptual integration. This thesis distinguishes emergent perceptual integration from non-emergent perceptual integration, with the former but not the latter resulting in visual descriptions unique to the whole and cannot be found in the parts; this distinction provides clues to the nature of unconscious processing (Chapter 1). Indeed, emergent properties, elusive as they sound, are tractable as revealed by emergent filling-in and its unique adaptation aftereffect (Chapter 2). To further understand the principles that constrain the emergence of visual descriptions, the thesis considers how spatial and temporal structures, two main types of contextual effects, regulate the emergence process. Rich contextual regularities are embedded in spatial configurations; a moving frame technique reveals that the coupling of objects to the contextual frames is pervasive and relatively automatic (Chapter 3), resulting in automatic deployments of exogenous visual attention to non-retinotopic, frame-centered locations (Chapter 4). Statistical regularities also abound in the vast associative knowledge one learns in the life time. Indeed, associated links can be established rapidly and without rewards, resulting in subsequent inventory attentional capture to the associated colors when the location of the target is uncertain (Chapter 5), and subsequent attentional inhibition of the associated colors when the location of the target is fixed (Chapter 6). These results thus reveal the importance in distinguishing emergent from non-emergent perceptual integration, and show that the emergence of visual descriptions is strongly constrained by spatial and temporal structures.