Scene images and text information‐based object location of robot grasping

With the rapid development of artificial intelligence (AI), application this technology in medical field is becoming increasingly extensive, along with a gradual increase amount intelligent equipment hospitals. Service robots can save human resources and replace nursing staff to achieve some work. In view phenomenon mobile service robots' grabbing distribution patients' drugs hospitals, real-time object detection positioning system based on image text information proposed, which realizes precise tracking objects completes grasping specific (medicine bottle). The lightweight model NanoDet used learn features category, bounding boxes are regressed. Then, images enhanced overcome unfavourable factors, such as small region. recognition PP-OCR detect recognise extract information. provided by two models fused, result matched box object. kernel correlation filter (KCF) algorithm introduced precisely control robot's grasping. Both deep learning adopt networks facilitate direct deployment. experiments show that proposed robot has high reliability, accuracy performance.