Develop Reference

Opencv Android : how to remove black background from thresholding output?

I am using OpenCV android library thresholding method for image segmentation, but the problem is that the output bitmap contains black background which I do not want please note that original image does not have any black background it is actually white. I am attaching the code for your reference, I am new to opencv and don't have much understanding about it also so kindly help me out.
private void Segmentation() {
Mat srcMat = new Mat();
gray = new Mat();
Utils.bitmapToMat(imageBmp, srcMat);
Imgproc.cvtColor(srcMat, gray, Imgproc.COLOR_RGBA2GRAY);
grayBmp = Bitmap.createBitmap(imageBmp.getWidth(), imageBmp.getHeight(), Bitmap.Config.RGB_565);
Utils.matToBitmap(gray, grayBmp);
grayscaleHistogram();
Mat threshold = new Mat();
Imgproc.threshold(gray, threshold, 0, 255, Imgproc.THRESH_BINARY_INV + Imgproc.THRESH_OTSU);
thresBmp = Bitmap.createBitmap(imageBmp.getWidth(), imageBmp.getHeight(), Bitmap.Config.RGB_565);
Utils.matToBitmap(threshold, thresBmp);
Mat closing = new Mat();
Mat kernel = Mat.ones(5, 5, CvType.CV_8U);
Imgproc.morphologyEx(threshold, closing, Imgproc.MORPH_CLOSE, kernel, new Point(-1, -1), 3);
closingBmp = Bitmap.createBitmap(imageBmp.getWidth(), imageBmp.getHeight(), Bitmap.Config.RGB_565);
Utils.matToBitmap(closing, closingBmp);
result = new Mat();
Core.subtract(closing, gray, result);
Core.subtract(closing, result, result);
resultBmp = Bitmap.createBitmap(imageBmp.getWidth(), imageBmp.getHeight(), Bitmap.Config.RGB_565);
Utils.matToBitmap(result, resultBmp);
Glide.with(ResultActivity.this).asBitmap().load(resultBmp).into(ivAfter);
}
enter image description here

What exactly do you want it to be then? Binary thresholding works like this:
if value < threshold:
value = 0
else:
value = 1
Of course you can convert it to a grayscale / RGB image and adjust the background to your liking. You can also invert your image (white background, black segmentation) by using the ~ operator.
segmented_image = ~ segmented_image
Edit: OpenCV has a dedicated flag to invert the results: CV_THRESH_BINARY_INV You are already using it, maybe try changing it to CV_THRESH_BINARY

Edge detection on colored background using OpenCV

I am using following code to detect edges from given document.
private Mat edgeDetection(Mat src) {
Mat edges = new Mat();
Imgproc.cvtColor(src, edges, Imgproc.COLOR_BGR2GRAY);
Imgproc.GaussianBlur(edges, edges, new Size(5, 5), 0);
Imgproc.Canny(edges, edges, 10, 30);
return edges;
}
And then I can find the document from this edges by finding largest contour from this.
My problem is I can find the document from following pic:
but not from following pic:
How can I improve this edge detection?

I use Python, but the main idea is the same.
If you directly do cvtColor: bgr -> gray for img2, then you must fail. Because the gray becames difficulty to distinguish the regions:
Related answers:
How to detect colored patches in an image using OpenCV?
Edge detection on colored background using OpenCV
OpenCV C++/Obj-C: Detecting a sheet of paper / Square Detection
In your image, the paper is white, while the background is colored. So, it's better to detect the paper is Saturation(饱和度) channel in HSV color space. For HSV, refer to https://en.wikipedia.org/wiki/HSL_and_HSV#Saturation.
Main steps:
Read into BGR
Convert the image from bgr to hsv space
Threshold the S channel
Then find the max external contour(or do Canny, or HoughLines as you like, I choose findContours), approx to get the corners.
This is the first result:
This is the second result:
The Python code(Python 3.5 + OpenCV 3.3):
#!/usr/bin/python3
# 2017.12.20 10:47:28 CST
# 2017.12.20 11:29:30 CST
import cv2
import numpy as np
##(1) read into bgr-space
img = cv2.imread("test2.jpg")
##(2) convert to hsv-space, then split the channels
hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
h,s,v = cv2.split(hsv)
##(3) threshold the S channel using adaptive method(`THRESH_OTSU`) or fixed thresh
th, threshed = cv2.threshold(s, 50, 255, cv2.THRESH_BINARY_INV)
##(4) find all the external contours on the threshed S
cnts = cv2.findContours(threshed, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)[-2]
canvas = img.copy()
#cv2.drawContours(canvas, cnts, -1, (0,255,0), 1)
## sort and choose the largest contour
cnts = sorted(cnts, key = cv2.contourArea)
cnt = cnts[-1]
## approx the contour, so the get the corner points
arclen = cv2.arcLength(cnt, True)
approx = cv2.approxPolyDP(cnt, 0.02* arclen, True)
cv2.drawContours(canvas, [cnt], -1, (255,0,0), 1, cv2.LINE_AA)
cv2.drawContours(canvas, [approx], -1, (0, 0, 255), 1, cv2.LINE_AA)
## Ok, you can see the result as tag(6)
cv2.imwrite("detected.png", canvas)

In OpenCV there is function called dilate this will darker the lines. so try the code like below.
private Mat edgeDetection(Mat src) {
Mat edges = new Mat();
Imgproc.cvtColor(src, edges, Imgproc.COLOR_BGR2GRAY);
Imgproc.dilate(edges, edges, Imgproc.getStructuringElement(Imgproc.MORPH_RECT, new Size(10, 10)));
Imgproc.GaussianBlur(edges, edges, new Size(5, 5), 0);
Imgproc.Canny(edges, edges, 15, 15 * 3);
return edges;
}

How to set Region of Interest around bounding rectangle in android using opecv

I have detected contour in my image. After detecting contour, I filtered contours based on area. Then I have drawn a bounding rectangle around license plate,around only one contour.
How to crop the remaining image and get only the region of rectangle drawn, that is, the I want to get only license plate around which I have drawn a rectangle. I dont have have coordinates of rectangle. I simply used Core.rectangle() function in opencv to draw a rectangle.
Can anyone help me with this. Please provide code in android + opencv using which i can ge desired results. I am also attaching an image.Here I have drawn rectangle using Core.rectangle(). I want to get the rectangle part and crop other part.

List<MatOfPoint> contour = new ArrayList<MatOfPoint>();
Mat hierarch = new Mat();
//find contours
Imgproc.findContours(gray1, contour, hierarch, Imgproc.RETR_LIST, Imgproc.CHAIN_APPROX_SIMPLE);
//itrerate through each contour
for (int Idx = 0; Idx < contours.size(); Idx++) {
double area = Imgproc.contourArea(contours.get(contourIdx));
if (area > largest_area) {
largest_area = (int) area1;
largest_contour_index = Idx;
Rect bounding_rect = Imgproc.boundingRect(contour.get(Idx));
Mat img = ROI.submat(bounding_rect1);
Bitmap resultBitmap = Bitmap.createBitmap(gray.cols(), gray.rows(), Bitmap.Config.ARGB_8888);
Utils.matToBitmap(gray, resultBitmap);
((ImageView) findViewById(R.id.imageView)).setImageBitmap(resultBitmap);
}
}

Android edge detection opencv

I am creating a project, where I have to remove the background from the image and detect the object.
I am using canny edge detection for detecting edges and than finding contours and than draw contours on a masked image, but after canny edge detection, I am getting broken edges ,how to fix that.
For Canny edge detection, for Threshold parameter, I have tried using thresholding with otsu's method for higher and lower threshold, but it doesn't seem to give appropriate result. Further, I have tried finding the mean of pixel values, and finding
double high_threshold = 1.33 * d;
double low_threshold = 0.66 * d;
it is also not giving accurate result. what else I can do
Mat rgba = new Mat();
Utils.bitmapToMat(bitmap, rgba);
Mat edges = new Mat(rgba.size(), CvType.CV_8UC1);Imgproc.cvtColor(rgba, edges, Imgproc.COLOR_RGB2GRAY, 4);
Imgproc.GaussianBlur(edges, edges, new Size(3,3), 2); Mat thresh=new Mat();
double upper_threshold = Imgproc.threshold(edges,thresh,0,255, Imgproc.ADAPTIVE_THRESH_GAUSSIAN_C| Imgproc.THRESH_OTSU);
double lower_threshold = 0.1*upper_threshold;Imgproc.Canny(edges,edges,upper_threshold,lower_threshold,3,false);Mat mDilatedMat = new Mat();
Mat Meroded = new Mat();
double erosion_size=5;
double dilation_size=4;
Mat e= Imgproc.getStructuringElement(Imgproc.MORPH_RECT, new Size(2*erosion_size + 1, 2*erosion_size+1));
Mat f= Imgproc.getStructuringElement(Imgproc.MORPH_RECT, new Size(2*dilation_size + 1, 2*dilation_size+1));
Imgproc.dilate(edges, mDilatedMat,e);
Imgproc.erode(mDilatedMat, Meroded,f);

You can improve your image extracted by sobel , canny or a different algorithm by applying edge linking algorithm.
Many edge linking algorithms are avaliable to use such as hough transform,
ant colony algorithm etc.

Detecting circles of a color gives "image must be 8-bit single-channel in HoughCircle" error in OpenCV

I am building an Android Application and I want to be able to detect black circles. I am using OpenCV3 for Android and I am able to filter out the black colours from my camera feed using the code below. Kindly note that I used the Color-blob-detection example given in the Android SDK and tweaked the code as such:
public Mat onCameraFrame(CameraBridgeViewBase.CvCameraViewFrame inputFrame) {
mRgba = inputFrame.rgba();
if (mIsColorSelected) {
Bitmap resultBitmap;
resultBitmap = Bitmap.createBitmap(mRgba.cols(), mRgba.rows(),Bitmap.Config.ARGB_8888);
Utils.matToBitmap(mRgba, resultBitmap);
//TODO - look for circles
Mat mat = new Mat(resultBitmap.getWidth(), resultBitmap.getHeight(), CvType.CV_8UC1);
Utils.bitmapToMat(resultBitmap, mat);
final Bitmap bitmap;
//TODO - filter out the black only
Mat mHSV = mat;
Mat mHSVThreshed = mat;
Mat mRgba2 = mat;
Imgproc.cvtColor(mat, mHSV, Imgproc.COLOR_BGR2HSV, 0);
Core.inRange(mHSV, new Scalar(0, 0, 0), new Scalar(130, 130, 130), mHSVThreshed);
Imgproc.cvtColor(mHSVThreshed, mat, Imgproc.COLOR_GRAY2BGR, 0);
Imgproc.cvtColor(mat, mRgba2, Imgproc.COLOR_BGR2RGBA, 0);
Imgproc.GaussianBlur(mRgba2, mRgba2, new Size(9, 9), 2, 2);
//this is for displaying purposes only.
//At this point, the image would be black and white, where the white spots are the black detected blobs
// Bitmap bmp = Bitmap.createBitmap(mRgba2.cols(), mRgba2.rows(), Bitmap.Config.ARGB_8888);
//Utils.matToBitmap(mRgba2, bmp);
//bitmap = bmp; //resultBitmap;
//TODO - new circle detection code: this uses the colour filtered Mat
mat = mRgba2;
Imgproc.HoughCircles(mat, circles,
Imgproc.CV_HOUGH_GRADIENT, 1, minDist, 100,
20, 0, 0);
/* get the number of circles detected */
int numberOfCircles = (circles.rows() == 0) ? 0 : circles.cols();
/* draw the circles found on the image */
for (int i = 0; i < numberOfCircles; i++) {
/* get the circle details, circleCoordinates[0, 1, 2] = (x,y,r)
* (x,y) are the coordinates of the circle's center
*/
double[] circleCoordinates = circles.get(0, i);
int x = (int) circleCoordinates[0], y = (int) circleCoordinates[1];
Point center = new Point(x, y);
int radius = (int) circleCoordinates[2];
/* circle's outline */
Imgproc.circle(mRgba2, center, radius, new Scalar(0,
200, 255), 4);
/* circle's center outline */
Imgproc.rectangle(mRgba2, new Point(x - 5, y - 5),
new Point(x + 5, y + 5),
new Scalar(0, 200, 255), -1);
}
Utils.matToBitmap(mRgba2, resultBitmap);
bitmap = resultBitmap;
runOnUiThread(new Runnable() {
#Override
public void run() {
mOpenCvCameraView.disableView();
mOpenCvCameraView.setVisibility(SurfaceView.GONE);
imageView.setVisibility(View.VISIBLE);
imageView.setImageBitmap(bitmap);
}
});
}
return mRgba;
}
What my code does is that it takes a snapshot of the camera feed as a Mat, and then using that Mat, I do some image processing on it and to black out everything except the black colour. The resulting Mat is the mRgba2 variable and I converted to Bitmap and displayed on an ImageView. I displayed this Bitmap to confirm that I am getting the result I want.
After I know that I'm able to filter out the colour I wanted, I then run a GaussianBlur on it then proceed to run HoughCircles. However, when I run the
Imgproc.HoughCircles(mat, circles,
Imgproc.CV_HOUGH_GRADIENT, 1, minDist, 100,
20, 0, 0);
line, I get an
The source image must be 8-bit, single-channel in function CvSeq* cvHoughCircles(CvArr*, void*, int, double, double, double, double, int, int)
Error.
I know that the code wherein I run HoughCircles on a given mat variable works, because I tested it before. Now, changing the mat variable I feed onto it does not, and I wonder what I did differently for the code to not work.

HoughCircles runs only on grayscale (CV8U) images.
Replacing
Imgproc.cvtColor(mat, mRgba2, Imgproc.COLOR_BGR2RGBA, 0);
with
Imgproc.cvtColor(mat, mRgba2, Imgproc.COLOR_BGR2GRAY, 0);
should remove the error.
Then the problem is to detect circles of a given color. This won't pose much of a problem in your case, because you are detecting black circles anyway. Instead of black circles, this will detect dark circles.
If you have dark colors other than black in some of the circles, you can filter them out separately by looking at the pixel values in a different color space.