Sony Pregius

Table of Contents

• Overview
  • Features
  • Supported pixel formats
  • Frame rate
• Special features
  • Compression

Overview

The following Sony Pregius S sensor models are available for the VisionCam XM2 or VisionSensor PV3 / PV4:

Supported sensors

Sensor	Type	Platform	eFG_CAMERA_TYPE	Number of usable pixels	Pixel size	Optical format
IMX565-AAMJ	Monochrome	VisionCam XM2	FG_CAMERA_TYPE_4096_3008_IMAGO_VCx_IMX565M	4096 x 3008	2.74 µm x 2.74 µm	1/1.1"
IMX565-AAQJ	Color		FG_CAMERA_TYPE_4096_3008_IMAGO_VCx_IMX565C
IMX567-AAMJ	Monochrome		FG_CAMERA_TYPE_2432_2064_IMAGO_VCx_IMX567M	2432 x 2064		1/1.8"
IMX567-AAQJ	Color		FG_CAMERA_TYPE_2432_2064_IMAGO_VCx_IMX567C
IMX568-AAMJ	Monochrome	VisionSensor PV3 / PV4	FG_CAMERA_TYPE_2472_2064_IMAGO_VSx_IMX568M	2472 x 2064
IMX568-AAQJ	Color		FG_CAMERA_TYPE_2472_2064_IMAGO_VSx_IMX568C
IMX900-AMR-C	Monochrome		FG_CAMERA_TYPE_2064_1552_IMAGO_VSx_IMX900M	2064 x 1552	2.25 µm x 2.25 µm	1/3.1"
IMX900-AQR-C	Color		FG_CAMERA_TYPE_2064_1552_IMAGO_VSx_IMX900C

Note

The SoC's MIPI CSI interface on the VisionSensor PV3 / PV4 requires at least two image buffers in the aquisition queue to avoid dropped sensor frames. A third buffer is required at the end of a frame, so that there are still remaining at least two buffers after one is given back to the user.

In order to avoid any dropped frames during startup while adding the first buffers with FG_append_image(), the sensor aquisition start is delayed until the second buffer is added. A third buffer should be added before the first frame completes.

Features

• Global shutter
• Overlapped integration and readout in free-run mode
• 10-bit to 8-bit conversion using linear mapping or compression (IMX56x only)
• Automatic black level correction
• 2x2 pixel binning
• Vertical and horizontal mirroring
• Short trigger delay < 1 µs
• Support for Line scan / multi-scan mode when using the VisionSensor PV3

Supported pixel formats

The supported pixel formats are listed below:

Supported pixel types

Sensor	eFG_PIXEL_TYPE	Output format
IMX56x-AAMJ, MX900-AMR-C (Monochrome)	FG_PIXEL_TYPE_Y_8	1 byte per pixel
IMX56x-AAQJ, MX900-AQR-C (Color)	FG_PIXEL_TYPE_Y_8	Raw bayer, 1 byte per pixel Note Adjusting color gain by using FG_set_gain_rgb() is not supported.
	FG_PIXEL_TYPE_RGB_24	3 bytes per pixel containing 8-bit R/G/B Note Available on the VisionSensor PV4 only and used as default. Uses the i.MX 95 NEO ISP pipeline for color conversion. FG_set_gain_rgb() can be used to apply digital color gain in the NEO ISP pipeline. The order of RGB components can be adjusted with FG_set_pixel_order() The AOI width must be a multiple of 64 pixels.

Frame rate

The frame rate can be increased be reducing the AOI height or by using binning mode. Use the calculator to obtain the maximum frame rate in free-run mode:

IMX565 Binning 2x2: AOI width: AOI height: Frame rate:

IMX567 Binning 2x2: AOI width: AOI height: Frame rate:

IMX568 Binning 2x2: AOI width: AOI height: Frame rate:

IMX900 Binning 2x2: AOI width: AOI height: Frame rate:

Note

The exposure duration reduces the achievable frame rate in triggered mode, because integration and readout of the previous frame cannot run simultaneously.

Special features

The sensor supports the following special features:

Special features

Property name	Description
Mirror	Flips the image horizontally and / or vertically 0: no flipping (default) 1: flip horizontally 2: flip vertically 3: flip horizontally and vertically (rotation by 180°) Note The order of color components for raw Bayer format depends on this setting. Use FG_get_pixel_order() to get the order after the Mirror option is configured.
Binning	Binning mode 1: no pixel binning (default) 2: 2x2 pixel binning Note Binning 2x2 is not available for color sensors. The binning mode must be configured before the first buffer is allocated. The parameters for FG_set_scan_param() and FG_set_aoi() refer to the reduced image size after binning is enabled. For IMX56x: binning increases pixel brightness by a factor of four (sum of four pixels). For IMX900: binning doesn't change the pixel brightness (average of four pixels).
ImageMode	Controls the digital test pattern: 0: Video output 1: Enable test pattern
FrameRate	Sets the desired sensor frame rate in Hertz. This setting is only applied in free run mode. The actual frame rate may be lower if the exposure time setting is too high, or if the sensor's limit is exceeded. 0: Use maximum frame rate ≥ 1: Desired frame rate in Hertz
ScanPeriod	Sets the desired sensor frame period in microseconds (similar to "FrameRate"). This setting is only applied in free run mode. The actual frame period may be longer if the exposure time setting is too high, or if the sensor's limit is exceeded. 0: Use maximum frame rate ≥ 1: Desired frame period in µs
Compression	Controls the mapping of 10 bit pixels to 8 bit output values, see Compression (not available for IMX900). 0: Linear mapping (default) 1: Compression curve 1 2: Compression curve 2
Temperature	Returns the sensor's temperature in °C Note The acquisition must have started to read the temperature.

Additional features are described in the platform pages for the VisionCam XM2 and VisionSensor PV3 / PV4.

Compression

The IMX56x sensors use a linear mapping of 10-bit pixel values to 8-bit output values by default. Compression can be enabled to provide higher resolution at low light and lower resolution at high light.

Three Compression settings are supported (0...2):

Note

Compression is not supported by the IMX900.