In Brazil, there are more than 6.5 million people with some visual
impairment. The total or partial loss of vision can cause a significant
impact on the lives of these people. The cane is now the most
widely available resource accessible to assist the locomotion of
these individuals. However, it is a limited resource. Once it cannot
classify the objects found, it only detects their existence. This work
contributes to new discussions on location and online mapping
on the theme of Assistive Technologies. We developed a system
that enables users to classify environments, promoting localization
through context science. Our solution uses images captured by a
webcam, computer vision techniques, and probabilistic analysis.
We evaluated on objective metrics showing satisfactory results.