How to improve line detection with OpenCV - opencv

Question

I am trying to detect lines in an image using OpenCV (Emgu.Net implementation).
So far, results are not that bad, but it appears to me not be consistent.
For instance, as you can see from the original and the results, some lines are well captured, yet the small top vertical ones are not, despite the fact that they have the same thickness.
Here is my code:
public static void ProcessImage(Mat src, bool createComposedImage)
{
//Convert the image to grayscale
Mat gray = new Mat();
CvInvoke.CvtColor(src, gray, ColorConversion.Bgr2Gray);
//Use Gaussian filtering to remove noise
CvInvoke.GaussianBlur(gray, gray, new Size(3, 3), 1);
double cannyThreshold = 180.0;
Mat result = new Mat();
List<Mat> resultPortions = new List<Mat>();
if (createComposedImage)
{
resultPortions.Add(src);
}
using (Mat cannyEdges = new Mat())
{
double cannyThresholdLinking = 120.0;
CvInvoke.Canny(gray, cannyEdges, cannyThreshold, cannyThresholdLinking);
LineSegment2D[] lines = CvInvoke.HoughLinesP(
image: cannyEdges,
rho: 1, //Distance resolution in pixel-related units
theta: Math.PI / 180.0, //Angle resolution measured in radians.
10, //threshold
30, //min Line width
10); //gap between lines
if (createComposedImage)
{
Mat lineImage = ProcessLinesInImage(gray, lines);
resultPortions.Add(lineImage);
}
if (createComposedImage)
{
try
{
CvInvoke.VConcat(resultPortions.ToArray(), result);
}
catch
{
}
foreach (Mat resultPortion in resultPortions)
{
resultPortion.Dispose();
}
}
}
return;
}
Here are the images