**5. Conclusions and Suggestions**

A construction job site covers the building footprint, work area, or material storage. With the simultaneous recognition of objects, such as workers, machines, and materials using a single shot multibox detector (SSD) in this case, it was found that the recognition performed better for large machines, including excavators, cranes, dump trucks, and concrete mixer trucks, with recognition accuracy close to 70%. Recognition accuracy was 53% for workers, and rebar was the least accurately identified of the three.

This study used the single shot multibox detector model with the VGG-16 neural network as its backbone network and VGG-16 is a 16-layer convolutional neural network, including 13 convolutional layers and 3 fully connected layers. A total of 320 construction

site construction images (80%) were trained, and the results could mark personnel, machinery, and materials simultaneously. The complexity of each on-site construction image was different; therefore, the time required for each image recognition was also different, but the average single image recognition time was 6 s. The object detection process encountered the following problems:


Based on the above, this study proposes future research directions regarding technology application, database construction, and algorithm optimization to enhance the accuracy and applicability of detection items:


Construction engineering is characterized by complexity; therefore, image recognition technology at construction sites enhances the safety and efficiency of construction site management. This technology enables more detailed identification and improvement of production efficiency and quality in the construction industry, thereby providing more significant development opportunities for the future of construction engineering.

**Author Contributions:** Conceptualization, Y.-R.W.; Methodology, L.-W.L.; Investigation, L.-W.L.; Resources, L.-W.L.; Writing—original draft, L.-W.L.; Writing—review & editing, L.-W.L.; Supervision, Y.-R.W.; Project administration, Y.-R.W.; Funding acquisition, Y.-R.W. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Data Availability Statement:** The Image in the research content is all owned by private companies, so they cannot be published.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
