Enhancing Object Detection by Using Probabilistic Spatial-Semantic Knowledge
Enhancing Object Detection by Using Probabilistic Spatial-Semantic Knowledge is a scholarly work, published in 2017 in ''Journal of Computers''. The main subjects of the publication include probabilistic logic, object, feature engineering, Simultaneous localization and mapping, artificial intelligence, question answering, and computer science. Autonomous mobile robots that act in human living environment and perform complex tasks there must be able to obtain relevant information from the surrounding and reason about this knowledge.The system's ability to recognize objects and assign them semantic description is a vital capability for the robots in order to carry out such a complex task.Dealing with objects in the environment is not a trivial problem but rather a challenging one and detection systems relay only on information extracted from their noisy sensors are often not sufficient to recognize objects properly.Therefore, this paper presents a new approach for object recognition, in which probabilistic spatial-semantic knowledge is applied to improve the recognition result.