Currently, indoor renovation construction technology is developing towards intelligence and automation, creating an urgent need for mobile robots to participate in high-pollution and high-risk construction operations. To address the issues of insufficient robustness and accuracy in indoor construction robot positioning due to relatively small indoor spaces and weak signals, this paper proposes a monocular vision indoor construction positioning and distance measurement method. It designs an improved YOLOv5 target detection algorithm with Canny edge detection operator for distance measurement and positioning. Experiments were conducted in a specific indoor construction environment by measuring the distance between the cross point projected by the laser emitter on the construction robot and the cross point generated by the laser level. The experimental results indicate that the calculated values from the algorithm differ from the actual measurements by no more than 8 mm, achieving an accuracy of 96%. This demonstrates that the proposed method has good performance in indoor construction positioning.