The invention provides an object detection method and a device. The invention relates to the field of artificial intelligence, and particularly to computer vision. The method comprises the following steps: acquiring a to-be-detected image; performing convolution processing on the detection image to obtain initial image features of a to-be-detected object in the to-be-detected image; determining enhanced image features of the to-be-detected object according to the knowledge graph information; and finally, determining candidate boxes and classifications of the to-be-detected object according tothe initial image features and the enhanced image features of the to-be-detected object. Since the enhanced image features indicate semantic information of different object types corresponding to other objects associated with the to-be-detected object, the effect of the object detection method can be improved.

Object detection method and device, and computer storage medium

The invention provides an image information target recognition method based on a data-information-knowledge graph structure, belongs to the crossing field of distributed computation and the software engineering technology, and aims to enable a machine to perform image recognition automatically and intelligently and find information which cannot be observed directly from an image. The structure based on a data graph, an information graph and a knowledge graph is established. Firstly, the structure based on the data graph, the information graph and the knowledge graph is established, then the three graphs are traversed on the basis of known entities in the image, the corresponding route is found from the graphs according to the image state, the position relation and the direction relation, new entities or relations (represented by multi-tuples) are obtained from the route by analyzing, hidden information is obtained, and information recognition is realized.

Shiqing He

Papers

Image information target recognition method based on data-information-knowledge graph structure