Karolliny Danielle Santos, Brenda Abreu Almeida, Caio Henrique Braga Teixeira, Guilherme Tadeu Almeida Sardinha Marques Guilherme Tadeu Almeida Sardinha Marques, Héber Stavrakas Gaipo, Pedro Henrique Pereira
This article explores applications of differential and integral calculus in computer vision, focusing on object detection in images. Using the MobileNet SSD architecture, the study demonstrates how calculus concepts such as derivatives, integrals, and optimization are fundamental to the development and implementation of image detection algorithms. The methodology, characterized as applied research with an action-research approach, involved a literature review, model selection, computational implementation of the code, and analysis of the results. The experiments demonstrated the efficiency of MobileNet SSD in object detection, processing times comparable to human reaction times. It is concluded that the underlying calculus is crucial for developing innovative and efficient solutions in computer vision.