Mast Arm Installation

Mast Arm Installation

DNNCam Mast Arm Installation Guide

This guide will walk through properly mounting a DNNCam on a horizontal mast arm.




Setup

A proper DNNCam installation requires:
- DNN-CAM-MNT - Mount Suitable for Mast Arm installation from Boulder AI. This includes the SE-0357-P33 Hub Plate from Pelco.
- Cable Gland for Ethernet cable - these can be provided by Boulder AI.
- Steel straps.
- A drip loop on the ethernet line.

The DNNCam silver 1-1/2" NPS compatible Mount will be shipped partially assembled. For remaining assembly, please reference the drawing below. A washer and lock washer must be used when attaching the mount to the mounting holes on the bottom of the DNNCam. This silver Mount will then screw into the Pelco SE-0357-P33 Hub Plate, as shown below.


Mounting Position and Orientation

In order to achieve the best detection of vehicles, bicycles, and pedestrians, the DNN Cam should be mounted on the traffic light pole arm as close to the center of the intersection as possible. This is illustrated in the figure below, where the red box illustrates the lateral region to permanently mount the DNN Cam to the traffic light pole arm.


The DNN Cam should be angled downward towards the opposite side of the intersection. The overall field of the view of the device should start near the center of the intersection and extend well past the crosswalk of the opposite side of the intersection. The figure below demonstrates the ideal mounting location of the DNN Cam on the traffic light pole arm and the approximate field of view.


The DNN Cam should be oriented such that the crosswalk is in the bottom third of the image. Because it may not be possible to see the live video stream from the DNN Cam during the initial installation on the traffic light pole arm, it is recommended to manually sight the device. From the rear of the device, look through the space between the silver visor and the black aluminum body of the DNN Cam. This space can be used as an approximation of what will be viewable to the DNN Cam, as illustrated in the image below. As illustrated, the left (red) view window is approximately the left-center of the image from the DNN Cam, while the right (blue) view window is approximately the right-center of the image.


Note: This method of setting up the camera FOV is not perfect. If possible, have somebody assist installation pull up the Web UI Liveview or RTSP stream from the camera to get a live look at the FOV of the DNNCam. This can prevent future adjustments to the camera in the case that the setup is less than ideal.




    • Related Articles

    • How to view RTSP streams from the DNNCam

      This guide will walk through accessing the RTSP streams produced by the DNNCam. Exporting an RTSP stream from the DNNCam  If your device is running Core Services v4.4 or earlier, then there is no option to export an RTSP stream in the Web UI. In this ...
    • Cable Gland Assembly Guide

      This article will explain how to set up and use the cable gland sold by Boulder AI to ensure that the connection to your DNN-Device remains safe in all weather conditions.  Cable glands are used in outdoor installations to keep the network connector ...
    • DNNCam Web UI v1 Setup Guide

      DNNCam Web UI v1 Guide This guide will help you access the Boulder AI Camera Web UI to configure camera network, optical, video, and user settings. NOTE: Web UI v2 was introduced with the release of Core Services v4.2. If your device is running Core ...
    • DNNCam Web UI v2 Setup Guide

      This guide will give an overview of the Web UI v2 that can be used to configure camera network, optical, and video settings. NOTE: Web UI v2 was introduced with the release of Core Services v4.2. If your device is running a Core Services version ...
    • Image Setting Overview

      This article will explain how each of the image adjustment settings in the Web UI will ultimately affect the image that the DNNCam produces. Having a clear and well adjusted image can help with detection and improve the accuracy of the AI ...