%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2017/08.21.23.26
%2 sid.inpe.br/sibgrapi/2017/08.21.23.26.52
%@doi 10.1109/SIBGRAPI.2017.35
%T Live monitoring in poultry houses: a broiler detection approach
%D 2017
%A Vilas Novas, Renan,
%A Usberti, Fábio Luiz,
%@affiliation Inst. of Comput., Univ. of Campinas (UNICAMP)
%@affiliation Inst. of Comput., Univ. of Campinas (UNICAMP)
%E Torchelsen, Rafael Piccin,
%E Nascimento, Erickson Rangel do,
%E Panozzo, Daniele,
%E Liu, Zicheng,
%E Farias, Mylène,
%E Viera, Thales,
%E Sacht, Leonardo,
%E Ferreira, Nivan,
%E Comba, João Luiz Dihl,
%E Hirata, Nina,
%E Schiavon Porto, Marcelo,
%E Vital, Creto,
%E Pagot, Christian Azambuja,
%E Petronetto, Fabiano,
%E Clua, Esteban,
%E Cardeal, Flávio,
%B Conference on Graphics, Patterns and Images, 30 (SIBGRAPI)
%C Niterói, RJ, Brazil
%8 17-20 Oct. 2017
%I IEEE Computer Society
%J Los Alamitos
%S Proceedings
%K Computer vision, object detection, broiler chickens, image processing, automatic monitoring.
%X This paper presents a general framework for live detection of broilers in poultry houses. The challenges for image recognition of broilers are posted by crowded scenes, poor image quality and difficulty in acquiring a benchmark of labeled samples. The proposed framework consists on the use of image thresholding, morphological transformations, feature engineering, in addition to supervised and unsupervised learn- ing techniques. Results show the effectiveness of the proposed framework to detect individual broilers in a poultry house image. Descriptive attributes related to the spatial distribution and movement of the broilers can be extracted using the resultant detections. These attributes can be used by automated warning systems, for the detection of anomalous events and thermal stress conditions.
%@language en
%3 PID4958805.pdf