Tuning the Content Detection Feature

This section describes how to tune the Content Detection feature to control the content detection submodule of LTM within a DC9400.

Before you begin

Make sure:

The details view contains the Content Detection table for LTM.
To tune this feature, you need to choose DC9400 > LTM > Content Detection in the Features view. For more details, see Tuning a DC9400 Feature.

Figure 1. Content Detection Table for LTM
A VTunerServer is accessible.
For information about how to set up a connection to a VTunerServer, see Setting Up a Connection to a VTunerServer.

About this task

This feature is used to tune the content detection submodule of LTM, which detects the existence of local details and generates accumulation weights for the local histogram. The detail detection is perfomed based on the differences of input luma and low-passed luma as follows:

$detail (Y) = Y - LPF (Y)$

where $LPF (Y) = convolve2D (Y, kernel)$ , $kernel = [\begin{matrix} Coef0 & Coef1 & Coef0 \\ Coef2 & Coef3 & Coef2 \\ Coef0 & Coef1 & Coef0 \end{matrix}]$ , and the coefficients need to be specified.

Procedure

Next to Content Detection Enable, select Enable to enable the content detection submodule of LTM.
Next to Content Detection Overlapping Enable, select Enable or Disable to control whether to allow overlapping for the content detection histogram.
In the Minimum Weight spin box, enter the minimum weight for content detection.
The supported value range is from 0 to 16, inclusive.
Next to Content Detection Coef, click View/Edit this Table and edit the 3x3 kernal coefficient table in the prompted dialog.
The sum of the coefficients must be less than 64.
You can use one of the following methods to update the Content Detection Coef table:
- Double-click and edit a single value in the table.
- Click Edit and update the values in the prompted Edit Table dialog.
  The values must be specified in the same order as those in the table and separated by a comma (,).
- Click Export to export the values to a .csv file, edit the file, and then import the updated values from the .csv file.
  The order of the values in the exported .csv file is the same as that in the table.
Next to Content Detection Threshold, click View/Edit this Table and edit the filtering ampliturde thresholds in the prompted dialog.
The columns in Content Detection Threshold are described as follows:
- Thresh0: If the filtering amplitude of a region is less than this threshold, the region is regarded as a flat area and is assigned with a smaller weight in histogram accumulation.
- Thresh1 and Thresh2: If the filtering ampliturde of a region is between these two thresholds, the region is regarded as a detail area and is assigned with a greater weight in histogram accumulation.
- Thresh3: If the filtering amplitude of a region is greater than this threshold, the region is regarded as an edge area and is assigned with a smaller weight in histogram accumulation.
The methods of updating the Content Detection Threshold table are similar to those of updating the Content Detection Coef table described in Step 4.
Next to Content Detection Slope, click View/Edit this Table and edit the slopes in the prompted dialog.
The methods of updating the Content Detection Slope table are similar to those of updating the Content Detection Coef table described in Step 4.
In the Filt Norm text box, enter the filter normalization value for content detection.
This value controls the depth of the content detection.

Results

The tuning results are printed in the Console view in real time.
The image displayed on the VTunerServer side is updated.