Boosting Frame Rate with ROI on Basler Cameras Using the pylon SDK (C# / .NET 8)

When developing machine-vision applications, you may find yourself asking:

“Can I increase the frame rate if I only need a portion of the image?”
“Can I reduce image size to speed up processing?”

The answer is often yes—by using ROI (Region of Interest).

By reducing the sensor readout area, you can significantly decrease data size and improve achievable frame rate. In this article, we explore how to configure ROI using the Basler pylon SDK in C#, along with real-world comparisons of full-frame vs ROI performance.

✔ What You Will Learn

How to configure ROI (Region of Interest) on Basler cameras
How ROI improves frame rate and processing speed
Common constraints, pitfalls, and error handling

✔ Environment

Item	Details
Camera	Basler acA2500-14gm
SDK	Basler pylon Camera Software Suite
Lang	C# / .NET 8 (Windows)
Lens	VS-LD25N (25 mm C-mount)

🔍 What Is ROI?

ROI (Region of Interest) defines a rectangular area on the camera sensor that you want to capture. Basler cameras expose four parameters to control ROI:

OffsetX — horizontal position
OffsetY — vertical position
Width — ROI width
Height — ROI height

Reducing ROI has several benefits:

Smaller image size
Shorter readout time → higher potential frame rate
Faster downstream processing

🔧 Setting ROI in C#

Below are helper methods added to the BaslerCameraSample class from previous articles:

public bool SetOffsetX(int offsetX)
    => SetPLCameraParamter(PLCamera.OffsetX, offsetX);

public bool SetOffsetY(int offsetY)
    => SetPLCameraParamter(PLCamera.OffsetY, offsetY);

public bool SetWidth(int width)
    => SetPLCameraParamter(PLCamera.Width, width);

public bool SetHeight(int height)
    => SetPLCameraParamter(PLCamera.Height, height);

public bool SetOffsetX(int offsetX)

=> SetPLCameraParamter(PLCamera.OffsetX, offsetX);

public bool SetOffsetY(int offsetY)

=> SetPLCameraParamter(PLCamera.OffsetY, offsetY);

public bool SetWidth(int width)

=> SetPLCameraParamter(PLCamera.Width, width);

public bool SetHeight(int height)

=> SetPLCameraParamter(PLCamera.Height, height);

Setting ROI in one operation

Because ROI must remain inside the sensor boundaries, the order of parameter updates matters.

ublic bool SetROI(int offsetX, int offsetY, int width, int height)
{
    if (width <= 0 || height <= 0)
        throw new ArgumentException(“ROI width and height must be greater than zero.”);

    if (GetWidth() + offsetX < GetSensorWidth())
    {
        if (!SetOffsetX(offsetX) || !SetWidth(width))
            return false;
    }
    else
    {
        if (!SetWidth(width) || !SetOffsetX(offsetX))
            return false;
    }

    if (GetHeight() + offsetY < GetSensorHeight())
    {
        if (!SetOffsetY(offsetY) || !SetHeight(height))
            return false;
    }
    else
    {
        if (!SetHeight(height) || !SetOffsetY(offsetY))
            return false;
    }

    return true;
}

public int GetWidth()
    => (int)GetPLCameraParamter(PLCamera.Width);

public int GetHeight()
    => (int)GetPLCameraParamter(PLCamera.Height);

public int GetSensorWidth()
    => (int)GetPLCameraParamter(PLCamera.SensorWidth);

public int GetSensorHeight()
    => (int)GetPLCameraParamter(PLCamera.SensorHeight);

ublic bool SetROI(int offsetX, int offsetY, int width, int height)

{

if (width <= 0 || height <= 0)

throw new ArgumentException(“ROI width and height must be greater than zero.”);

if (GetWidth() + offsetX < GetSensorWidth())

{

if (!SetOffsetX(offsetX) || !SetWidth(width))

return false;

}

else

{

if (!SetWidth(width) || !SetOffsetX(offsetX))

return false;

}

if (GetHeight() + offsetY < GetSensorHeight())

{

if (!SetOffsetY(offsetY) || !SetHeight(height))

return false;

}

else

{

if (!SetHeight(height) || !SetOffsetY(offsetY))

return false;

}

return true;

}

public int GetWidth()

=> (int)GetPLCameraParamter(PLCamera.Width);

public int GetHeight()

=> (int)GetPLCameraParamter(PLCamera.Height);

public int GetSensorWidth()

=> (int)GetPLCameraParamter(PLCamera.SensorWidth);

public int GetSensorHeight()

=> (int)GetPLCameraParamter(PLCamera.SensorHeight);

ROI using `Rect` (WPF)

using Windows;

public bool SetROI(Rect roi)
    => SetROI((int)roi.X, (int)roi.Y, (int)roi.Width, (int)roi.Height);

using Windows;

public bool SetROI(Rect roi)

=> SetROI((int)roi.X, (int)roi.Y, (int)roi.Width, (int)roi.Height);

📈 Why ROI Increases Frame Rate

Sensor readout time is almost always proportional to the number of pixels captured. Reducing the readout area:

Lowers pixel count
Reduces required bandwidth
Allows the camera to reach higher frame rates

This effect is independent of exposure time, so even long exposures benefit from faster readout afterward.

✔ Constraints When Setting ROI

Parameter	Notes
Width / Height	Many Basler models require values to follow a specific increment (2, 4, 8, 16… depending on sensor)
OffsetX / OffsetY	Must not exceed sensor boundaries

To inspect the increment value:

public long GetInc(IntegerName parameter)
{
    if (Camera == null || !IsConnected)
        throw new InvalidOperationException(“Camera is not connected.”);

    if (Camera.Parameters[parameter].IsReadable)
    {
        long inc = Camera.Parameters[parameter].GetIncrement();
        Console.WriteLine($”{parameter} increment: {inc}”);
        return inc;
    }

    throw new InvalidOperationException($”{parameter} is not readable.”);
}

public long GetInc(IntegerName parameter)

{

if (Camera == null || !IsConnected)

throw new InvalidOperationException(“Camera is not connected.”);

if (Camera.Parameters[parameter].IsReadable)

{

long inc = Camera.Parameters[parameter].GetIncrement();

Console.WriteLine($“{parameter} increment: {inc}”);

return inc;

}

throw new InvalidOperationException($“{parameter} is not readable.”);

}

Example test:

[TestMethod()]
public void GetIncTest()
{
    if (!_baslerCameraSample.IsConnected)
        _baslerCameraSample.Connect();

    var incWidth = _baslerCameraSample.GetInc(PLCamera.Width);
    Assert.AreEqual(1, incWidth);

    var incHeight = _baslerCameraSample.GetInc(PLCamera.Height);
    Assert.AreEqual(1, incHeight);
}

[TestMethod()]

public void GetIncTest()

{

if (!_baslerCameraSample.IsConnected)

_baslerCameraSample.Connect();

var incWidth = _baslerCameraSample.GetInc(PLCamera.Width);

Assert.AreEqual(1, incWidth);

var incHeight = _baslerCameraSample.GetInc(PLCamera.Height);

Assert.AreEqual(1, incHeight);

}

📏 Testing Frame Rate with ROI

static BaslerCameraSample _baslerCameraSample = new BaslerCameraSample();

[TestMethod()]
public void SetROITest()
{
    _baslerCameraSample.Connect();

    _baslerCameraSample.EnableFrameRate(true);
    _baslerCameraSample.SetFrameRate(30); // 30 fps target

    _baslerCameraSample.SetROI(1246, 1200, 640, 480);
    var fps1 = _baslerCameraSample.GetResultingFrameRate();
    _baslerCameraSample.BurstCapture(5, “ROITest_640x480”);

    _baslerCameraSample.SetROI(0, 0, 2592, 1944);
    var fps2 = _baslerCameraSample.GetResultingFrameRate();
    _baslerCameraSample.BurstCapture(5, “ROITest_2592x1944″);

    Console.WriteLine($”FPS with ROI: {fps1:F2}, full-frame FPS: {fps2:F2}”);
    Assert.IsTrue(fps1 > fps2);
}

static BaslerCameraSample _baslerCameraSample = new BaslerCameraSample();

[TestMethod()]

public void SetROITest()

{

_baslerCameraSample.Connect();

_baslerCameraSample.EnableFrameRate(true);

_baslerCameraSample.SetFrameRate(30); // 30 fps target

_baslerCameraSample.SetROI(1246, 1200, 640, 480);

var fps1 = _baslerCameraSample.GetResultingFrameRate();

_baslerCameraSample.BurstCapture(5, “ROITest_640x480”);

_baslerCameraSample.SetROI(0, 0, 2592, 1944);

var fps2 = _baslerCameraSample.GetResultingFrameRate();

_baslerCameraSample.BurstCapture(5, “ROITest_2592x1944”);

Console.WriteLine($“FPS with ROI: {fps1:F2}, full-frame FPS: {fps2:F2}”);

Assert.IsTrue(fps1 > fps2);

}

🧪 Real-World Comparison

Full Resolution (2592 × 1944)

Frame timestamps demonstrate slower readout:

Frame	0	1	2	3	4
T [ms]	72	140	209	277	346

ROI (640 × 480)

Frame	0	1	2	3	4
T [ms]	23	67	87	120	154

Summary Comparison

Mode	Resolution	Resulting Frame Rate
Full Frame	2592×1944	~14.6 fps
ROI	640×480	~30.0 fps

→ ROI doubles the effective frame rate in this test.

⚠️ Common Errors & Solutions

Issue	Cause
`InvalidParameterException`	ROI exceeds sensor bounds, or width/height violates increments
No frames after setting ROI	Some camera models require reinitialization after ROI changes

📝 Summary

ROI can significantly reduce readout time and increase frame rate
Combining ROI with proper frame-rate settings improves real-time performance
Pay attention to width/height increments and sensor boundaries
Ideal for speeding up burst capture and reducing processing load

筆者: @MilleVision

📦 Full Sample Code Package

👉 https://millevision.booth.pm/items/7316233 Includes ROI, burst capture, trigger control, and full pylon SDK integration.(With Japanese Comment)

Boosting Frame Rate with ROI on Basler Cameras Using the pylon SDK (C# / .NET 8)

Boosting Frame Rate with ROI on Basler Cameras Using the pylon SDK (C# / .NET 8)

✔ What You Will Learn

✔ Environment

🔍 What Is ROI?

🔧 Setting ROI in C#

Setting ROI in one operation

ROI using `Rect` (WPF)

📈 Why ROI Increases Frame Rate

✔ Constraints When Setting ROI

📏 Testing Frame Rate with ROI

🧪 Real-World Comparison

Full Resolution (2592 × 1944)

ROI (640 × 480)

Summary Comparison

⚠️ Common Errors & Solutions

📝 Summary

📦 Full Sample Code Package

いいね:

コメント

コメントを残すコメントをキャンセル

投稿をさらに読み込む

Software Trigger Capture with the Basler pylon SDK (C# / .NET)

Reproducing Camera Conditions: Saving and Loading Basler Settings as PFS Files

Displaying and Saving Basler Camera Images with OpenCV

Stabilizing Basler Camera Acquisition with Event-Driven Capture and Async Saving (C# / pylon SDK)

Boosting Frame Rate with ROI on Basler Cameras Using the pylon SDK (C# / .NET 8)

Boosting Frame Rate with ROI on Basler Cameras Using the pylon SDK (C# / .NET 8)

✔ What You Will Learn

✔ Environment

🔍 What Is ROI?

🔧 Setting ROI in C#

Setting ROI in one operation

ROI using Rect (WPF)

📈 Why ROI Increases Frame Rate

✔ Constraints When Setting ROI

📏 Testing Frame Rate with ROI

🧪 Real-World Comparison

Full Resolution (2592 × 1944)

ROI (640 × 480)

Summary Comparison

⚠️ Common Errors & Solutions

📝 Summary

📦 Full Sample Code Package

いいね:

コメント

コメントを残す コメントをキャンセル

投稿をさらに読み込む

Software Trigger Capture with the Basler pylon SDK (C# / .NET)

Reproducing Camera Conditions: Saving and Loading Basler Settings as PFS Files

Displaying and Saving Basler Camera Images with OpenCV

Stabilizing Basler Camera Acquisition with Event-Driven Capture and Async Saving (C# / pylon SDK)

ROI using `Rect` (WPF)

コメントを残すコメントをキャンセル