Correction Settings

The Correction tab of the Acquisition panel allows you to configure field correction. Although all vision system are calibrated during production to produce undistorted point clouds, it is important to perform additional correction to account for misalignment between the vision system and the motion direction. Field correction is performed based on the encoder resolution or motion speed, depending on the selected Motion Input.

You can configure field correction using one of the following field correction types:

Target Correction: When selecting this method, you must use a Cognex Correction Target Model to perform field correction. This method handles possible installation errors and also calculates the appropriate Encoder Resolution (in case of encoder-based image acquisition) or Motion Speed (in case of time-based image acquisition). Positive Encoder (+) and Acquisition Direction are also determined during this process. This method is very accurate, and is able to compensate all possible sensor misalignments.
Manual Correction: When selecting this method, you must specify the identified installation misalignments (in terms of rotational angles) yourself. Likewise, you must set the appropriate Encoder Resolution (in case of encoder-based image acquisition) or Motion Speed (in case of time-based image acquisition) values manually.

Note: Cognex recommends using Manual Correction either if you do not have any Cognex Correction Target Models, or because they cannot be used due to space limitations in the physical setup. After performing manual correction, Cognex recommends to Lock the correction to ensure that no settings affecting Field Correction will be modified accidentally later.

Note:

After Target Correction is performed, the Manual Correction settings are automatically updated, allowing you to check the values determined by the target correction algorithm. You can also modify these settings afterwards without the need of performing Target Correction again: to do so, just switch from Target Correction to Manual Correction, and modify the setting you want to change. For example, in case of time-based acquisition, you can change the Motion Speed setting, if needed – just consider that acquisition accuracy depends on how close actual motion is to the configured motion value.

Parameter	Description
Target Correction
Target Model	Allows you to select the Cognex Field Correction Target Model type used for correction. The model is defined based on the square edge length at the top of the frustum. Note: Select the target model based on the working space. Always use a model that covers a significant part of the working space. Make sure that the selected setting corresponds to the target model you are using for correction. For example, do not select TS10 in the menu if you are using a TS20 or TS40 target model.
Correction Mode	Allows you to select the mode for determining the correction parameters: Fast: This mode performs a single target scan (in any orientation) to perform field correction. Select Fast mode to quickly apply field correction, and when the inaccuracy of the target is not a concern. Robust: This mode performs three target scans in different orientations to perform field correction. Select Robust mode if you need to compensate manufacturing errors on the field correction target.
Estimated Scan Time [s]	Defines the estimated duration of the scan (in seconds) required for target correction. The value range is 0 – 500, and the default value is 2. Note: Specify a duration that is enough to cover the entire correction target end-to-end. If the estimated scan time is set too low, scanning will be incomplete, causing the field correction to fail. If it is set too long, the scan will have a low Y-resolution, making field correction less accurate.
Grab One and Perform Correction	Start the acquisition of the specified Field Correction Target. If acquisition is successful, the correction settings get locked. If any errors occur, it is indicated in a pop-up window with further instructions.
Manual Correction
Motion Speed [mm/s]	Sets the speed along the motion of the system (in mm/s), such as the speed of a conveyor belt, motion stage, or robot arm. Note: This setting is enabled only if Motion Input in the tab is set to Time, and Manual Correction is enabled. After field correction has been performed, the system corrects the value to the actual motion speed.
Encoder Resolution [µm]	Defines the resolution of the encoder used by the vision system (in µm), corresponding to the distance between consecutive pulse outputs. Note: This setting is enabled only if Motion Input in the tab is set to Encoder, and Manual Correction is enabled. After Target Correction, the value is corrected to the actual encoder resolution.
Positive Encoder (+)	Specifies how the encoder is physically connected to the vision system, affecting the motion direction that increases the Encoder Counter. Laser to Lens - the Encoder Counter increases if the motion direction is from the laser to the lens of the vision system. Lens to Laser - the Encoder Counter increases if the motion direction is from the lens of the vision system to the laser. If this setting is correctly configured, the Current Indicator correctly indicates the direction of movement. Otherwise, you must correct the value of the setting. This setting can only be configured before field correction is performed. After that, it is grayed out and can no longer be modified. Note: This setting is enabled only if Motion Input is set to Encoder, and Encoder Type is set to Dual Channel. • In case of target field correction, this setting is automatically determined, therefore you do not need to configure it in advance. Tip: Make sure that the Encoder Direction is specified correctly, based on the physical setup of the vision system. Otherwise, the acquisition occurs in the opposite direction compared to the requested Acquisition Direction.
Rotation [deg]	Specifies the angle between the vision system and motion direction on the XY plane. In other words, it is the rotation along the Z axis required to align the device and the motion vector. The value range is -45 to 45, and the default value is 180. Note: This parameter is calculated automatically if Target Correction is enabled.
Tilt [deg]	Specifies the angle between the laser plane and the Z axis. Use this setting if the laser is not perpendicular to the base plane. The value range is 0 to 180, and the default value is 180. Note: This parameter is calculated automatically if Target Correction is enabled.
Tilt Direction [deg]	Specifies the direction in which the device is tilted. This value is ignored if the tilt angle is 0. The value range is -180 to 180, and the default value is 180. Note: This parameter is calculated automatically if Target Correction is enabled.
Lock	Applies the configured correction settings, and locks the Correction tab. To modify any of the properties, unlock the Correction tab with the Unlock button. Note: You can export Manual Correction settings only if the settings are locked with the Lock button. At the same time, you cannot import any previously saved correction settings while the lock is active.
Shared Settings
Import...	Allows you to import an existing field correction configuration set for the configured vision system. Note: This button is enabled only if the correction settings are unlocked.
Export...	Allows you to export the field correction settings configured on the device, and reuse them on another vision system. The field correction configuration is saved into an .ar file. Note*: This button is enabled only if the correction settings are locked.

Once Field Correction is performed, the settings cannot be modified unless clicking the Unlock button, or performing a new correction. The results of the correction then appear under the Unlock button.

Note:

If the resulting RMS value of the correction is higher than expected, the Results table indicates it in red, and a pop-up warning also appears at the top right corner of the screen. This error can occur if you:

Use an incorrect frustum model.
Configure the Detection Zone incorrectly (for example, the floor is also detected).
Configure other acquisition settings incorrectly, so the frustum could not be acquired by the vision system.