To configure Water level detection, do the following:

1. To record water level detection readings to the archive, set Yes for the Record mask to archive parameter (1).
   

2. If the camera supports multistreaming, select the stream for which detection is needed. Select a low-quality video stream to reduce the Server load (2).
3. Select a processing resource for decoding video streams (3). When you select a GPU, a stand-alone graphics card takes priority (when decoding with NVIDIA NVDEC chips). If there is no appropriate GPU, the decoding will use the Intel Quick Sync Video technology. If neural network is not used, the algorithm can work only on CPU (see General Information on Configuring Detection).

4. Analyzed framed are scaled down to a specified resolution (4, 1920 pixels on the longer side). This is how it works:

   1. If the longer side of the source image exceeds the value specified in the Frame size change field, it is divided by two.

   2. If the resulting resolution falls below the specified value, it is used further.

   3. If the resulting resolution still exceeds the specified limit, it is divided by two, etc.

      For example, the source image resolution is 2048*1536, and the specified value is set to 1000. In this case, the source resolution will be halved two times (512*384), as after the first division, the number of pixels on the longer side exceeds the limit (1024 > 1000).

      
      

      If detection is performed on a higher resolution stream and detection errors occur, it is recommended to reduce the compression.

      
      

5. Set the frame rate value for the detection tool to process per second (5). The value should be in the range [0,016; 100].

6. If the water in the frame is transparent and the detection tool cannot correctly identify its level, use Neural network (6):

   1. If No is selected, the detection tool will work based on the algorithm without using the neural network, ignoring the value specified in the Neural network file field.

   2. If Yes is selected:

      1. If no value is selected in the Neural network file field, the system will automatically select the required standard neural network for the device specified in the Neural network mode field.

      2. If a custom neural network is selected in the Neural network file field that corresponds to the device specified in the Neural network mode field and is a water level neural network, the detection tool will create an engine using this network.
         

         If the neural network file is not specified correctly, the detection tool will not work. The engine will recreate itself every 20 seconds.

         
         

7. Select the neural network file (7). The following standard neural networks for different processor types are located in the C:\Program Files\Common Files\AxxonSoft\DetectorPack\NeuroSDK directory:
   

   WaterLevelRuleNet_movidius.ann	Water level detection / IntelNCS
   WaterLevelRuleNet_openvino.ann	Water level detection / CPU
   WaterLevelRuleNet_origin_onnx.ann	Water level detection / GPU

   Enter the path to a custom neural network file into this field.
   

   For correct neural network operation on Linux, place the corresponding file in the /opt/AxxonSoft/DetectorPack/NeuroSDK directory.

   
   

8. Select the processor for the neural network − CPU, one of GPUs, or one of Intel processors (8) (see Hardware requirements for neural analytics operation, General Information on Configuring Detection). 
   

   It may take several minutes to launch the algorithm on NVIDIA GPU after you apply the settings. You can use caching to speed up future launches (see Configuring the acceleration of GPU-based neuroanalytics).

   
   

   If you specify other processing resource than the CPU, this device will carry the most of computing load. However, the CPU will also be used to run the detection tool.

   
   

9. On the measurement scale (9, 10), set the top and bottom visible values in normal conditions. The value should be in the range [0; 19].
10. Move the anchor points in the preview window:

    1. Set the measurement scale in the frame.
       

       Top and bottom values of the measurement scale should match the actual settings (see i. 9).

       
       

       Water level sensor is shown in the lower left corner. If the sensor is blue, the water level is below high and critical marks. If the sensor is yellow, the water level is at high mark, but below critical mark. A red sensor means that water level is above critical mark.

       
       

    2. Draw a line to set the top limit for water level upon reaching which the detection tool triggers an alarm.
    3. Draw a line to set the top limit for water level upon reaching which the detection tool highlights the sensor icon in the video surveillance window with yellow color.
11. Click the Apply button.

Configuring Water level detection tool is complete.

When you have created a detection tool, you can see a sensor on the layout in the video surveillance window.

If the sensor icon is green , the water level is lower than both critical and high marks. If the icon is yellow , the water level is above the high mark, but below the critical mark. A red icon means that the water level is above the critical mark.

You can also add a numerical value of the water level to the video surveillance window (see Configuring display of water level detection).