The instrumental Brahmin and the "half-caste" computer: Astronomy and colonial rule in Madras, 1791-1835

What did science make possible for colonial rule? How was science in turn marked by the knowledge and practices of those under colonial rule? Here I approach these questions via the social history of Madras Observatory. Constructed in 1791 by the East India Company, the observatory was to provide local time to mariners and served as a clearinghouse for the company's survey and revenue administration. The astronomical work of Madras' Brahmin assistants relied upon their knowledge of jyotiśāstra [Sanskrit astronomy/astrology], and can be seen as a specialized form of the kind of South Indian scribal labor and knowledge that also staffed the company's tax offices. If at Greenwich the division of labor meant observatory work bore resemblances to the factory and the accounts office, in Madras, astronomy and accounting drew on similar labor forms because they were part of the same enterprise. But the company did not just adapt preexisting forms of labor, it also attempted to produce its own at a school built near the observatory to train "half-caste" orphans as apprentice surveyors and assistant computers. The school, staffed by the Brahmins, drew upon knowledge and pedagogical practice associated with the tinnai, the schools in which upper-caste children learned to read, write, and calculate. For a time, the observatory's social order was literally "half-caste." The paper also considers how the relationship between caste, status, and instrument was reflected in the visual and material culture of the observatory, such as in Indian-language inscriptions on its central pillar. For company astronomers, the measurement of time meant reworking the relationships among the Indian past, the colonial present, and an imperial posterity. Science under colonial rule spanned multiple temporal and social registers because it was the result of negotiations between the demands of political economy and the knowledge and practices of colonized others.

Learning channel-selective and aberrance repressed correlation filter with memory model for unmanned aerial vehicle object tracking

To ensure that computers can accomplish specific tasks intelligently and autonomously, it is common to introduce more knowledge into artificial intelligence (AI) technology as prior information, by imitating the structure and mindset of the human brain. Currently, unmanned aerial vehicle (UAV) tracking plays an important role in military and civilian fields. However, robust and accurate UAV tracking remains a demanding task, due to limited computing capability, unanticipated object appearance variations, and a volatile environment. In this paper, inspired by the memory mechanism and cognitive process in the human brain, and considering the computing resources of the platform, a novel tracking method based on Discriminative Correlation Filter (DCF) based trackers and memory model is proposed, by introducing dynamic feature-channel weight and aberrance repressed regularization into the loss function, and by adding an additional historical model retrieval module. Specifically, the feature-channel weight integrated into the spatial regularization (SR) enables the filter to select features. The aberrance repressed regularization provides potential interference information to the tracker and is advantageous in suppressing the aberrances caused by both background clutter and appearance changes of the target. By optimizing the aforementioned two jointly, the proposed tracker could restrain the potential distractors, and train a robust filter simultaneously by focusing on more reliable features. Furthermore, the overall loss function could be optimized with the Alternative Direction Method of Multipliers (ADMM) method, thereby improving the calculation efficiency of the algorithm. Meanwhile, with the historical model retrieval module, the tracker is encouraged to adopt some historical models of past video frames to update the tracker, and it is also incentivized to make full use of the historical information to construct a more reliable target appearance representation. By evaluating the method on two challenging UAV benchmarks, the results prove that this tracker shows superior performance compared with most other advanced tracking algorithms.