At present, industrial robots cannot achieve complete automatic assembly, for some specific occasions still need auxiliary facilities to cooperate. The identification and positioning of workpiece is the most important technical difficulty in automatic grasping of workpiece. This paper mainly studies the improvement and realization of the processing and positioning method of intelligent construction machinery components relying on machine vision. The overall scheme design of target workpiece positioning is presented. This paper introduces the technical background and significance of workpiece recognition and location based on machine vision. According to the actual engineering requirements of the workpiece, a monocular vision method of workpiece identification and positioning is designed. A localization algorithm based on image deep learning is designed. The deep learning network is used as the image feature extractor, and the extracted features are used to predict the corresponding Angle of the image by linear regression method. The experimental results show that the workpiece object detection method based on the improved deep learning model reduces the missing detection rate of small objects and improves the accuracy of the overall workpiece detection.