01863naa a2200337 a 450000100080000000500110000800800410001902400270006010000220008724501290010926000090023850000260024752008420027365000200111565000290113565000170116465000180118165000160119965300180121565300340123365300230126765300280129065300170131865300300133565300180136565300350138370000210141870000180143970000220145777300460147921216642020-04-17       2020    bl uuuu     u00u1 u    #d7 a10.3390/s200721262DOI1 aBARBEDO, J. G. A.  aCounting cattle in UAV imagesbdealing with clustered animals and animal/background contrast changes.h[electronic resource]  c2020  aArticle number: 2126.  aAbstract: The management of livestock in extensive production systems may be challenging, especially in large areas. Using Unmanned Aerial Vehicles (UAVs) to collect images from the area of interest is quickly becoming a viable alternative, but suitable algorithms for extraction of relevant information from the images are still rare. This article proposes a method for counting cattle which combines a deep learning model for rough animal location, color space manipulation to increase contrast between animals and background, mathematical morphology to isolate the animals and infer the number of individuals in clustered groups, and image matching to take into account image overlap. Using Nelore and Canchim breeds as a case study, the proposed approach yields accuracies over 90% under a wide variety of conditions and backgrounds.  aNeural networks  aUnmanned aerial vehicles  aGado Canchim  aGado de Corte  aGado Nelore  aCanchim breed  aConvolutional neural networks  aDeep learning mode  aMathematical morphology  aNelore breed  aRede neural convolucional  aRedes neurais  aVeículo aéreo não tripulado1 aKOENIGKAN, L. V.1 aSANTOS, P. M.1 aRIBEIRO, A. R. B.  tSensorsgv. 20, n. 7, p. 1-14, Apr. 2020.