A Cervix Detection Driven Deep Learning Approach for Cow Heat Analysis from Endoscopic Images

Ruiwen He (1, 2, 3) , Halim Benhabiles (1, 2, 3) , Feryal Windal (1, 2, 3) , Gael Even (4) , Christophe Audebert (4) , Dominique Collard (3, 5) , Abdelmalik Taleb-Ahmed (1, 6, 7)

1 IEMN - Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520
2 BIOMEMS - IEMN - Bio-Micro-Electro-Mechanical Systems - IEMN
3 JUNIA - JUNIA
4 Gènes Diffusion [Douai]
5 LIMMS - Laboratory for Integrated Micro Mechatronics Systems
6 COMNUM - IEMN - COMmunications NUMériques - IEMN
7 UPHF - Université Polytechnique Hauts-de-France

In this article, we propose a new approach for the cow heat detection from endoscopic images. Our approach permits to identify on the fly the cow heat state through two successive stages, namely cervix detection then heat classification. For this purpose, images are analyzed by a Transformer based detection model to localize the cervix, in which case they are analyzed by a CNN-based heat classification model. The proposed approach permits to assist the farmer during the insemination operation by localizing the cervix in an accurate way. Moreover, the confidence level of the final decision of the classification model is increased by focusing its analysis only on cervix images. The effectiveness of our method is demonstrated on our generated dataset and the obtained performance outperform the state of the art.

Mots clés: Deep learning; Transformer; Endoscopic images; Artificial insemination; Cervix detection

Domaines: Physique [physics]; Sciences de l'ingénieur [physics]

Ruiwen He, Halim Benhabiles, Feryal Windal, Gael Even, Christophe Audebert, et al.. A Cervix Detection Driven Deep Learning Approach for Cow Heat Analysis from Endoscopic Images. 2022 IEEE International Conference on Image Processing (ICIP), Oct 2022, Bordeaux, France. pp.3672-3676, ⟨10.1109/ICIP46576.2022.9897442⟩. ⟨hal-03839222⟩