I am using opencv.js to detect some lines in an image using the houghp transform. This is the result I am getting -

And here is the function call -

```
let lines = new cv.Mat();
cv.HoughLinesP(canny, lines, 1, Math.PI / 180, 50, 50, 5);
```

I want the lines that are parallel and close to each other to be only counted as one line. I tried changing the maxlinedistance but it is not reliable enough. I tried looping over the lines and comparing the x and y coordinates of the detected lines. The problem is, I couldn’t figure out how to delete the lines that I don’t need. Here is the code for that -

```
for (let i = 0; i < lines.rows; ++i) {
let x1 = lines.data32S[i * 4];
let y1 = lines.data32S[i * 4 + 1];
let x2 = lines.data32S[i * 4 + 2];
let y2 = lines.data32S[i * 4 + 3];
for (let j = i+1; j < lines.rows; ++j) {
let x3 = lines.data32S[j * 4];
let y3 = lines.data32S[j * 4 + 1];
let x4 = lines.data32S[j * 4 + 2];
let y4 = lines.data32S[j * 4 + 3];
if (Math.abs(x1-x3) < 10 || Math.abs(y1-y3) < 10 || Math.abs(x2-x4) < 10 || Math.abs(y2-y4) < 10) {
lines.row(j).delete(); // pseudocode
j--;
}
}
}
```

Any help is appreciated. Thanks!