Ask Your Question

Chris's profile - activity

2019-04-20 15:45:29 -0500 commented question How to remove boundary from binarized image

Look at Opencv's contour finding. You should be able to isolate just the digits by finding all the contours and filteri

2019-04-20 15:30:53 -0500 commented question Unable to load images with imread in Visual Studio 2019

Likely a path and/or extension issue. Try using an absolute path and check to see if the image file has an extension lik

2019-04-19 10:35:34 -0500 answered a question Process image to get rid of line between two components

If I do an open morph operation with MorphologyEx using a rect kernel that is 12 x 1, I get this image If I subtract

2019-04-17 07:25:48 -0500 answered a question Drawing a rectangle and masking the outside area

There is a version of the copyto function that takes a mask image. A mask image is the same size as the image copying f

2019-04-16 10:22:49 -0500 commented question I need help in drawing bounding rectangle on threshold Codebook algorithm

You could try contour finding (https://docs.opencv.org/4.0.0/dc/dcf/tutorial_js_contour_features.html) but not sure how

2019-04-15 07:40:10 -0500 commented question Read QR code and mark it as green if found in excel file

If you can save the excel table as csv file, then it is pretty easy to read as a text file.

2019-04-15 07:32:26 -0500 commented question Double digit recognition

If the digits don't vary too much in scale or rotation, we find template matching to work well.

2019-04-15 07:18:19 -0500 answered a question Detect decimal or small dot in image

Another solution may be to threshold the image, do a few iterations of dilation to join the pixels, then look for contou

2019-04-14 14:59:15 -0500 received badge  Student (source)
2019-04-14 14:40:27 -0500 asked a question What is the THRESH_MASK ThresholdType in cv4?

What is the THRESH_MASK ThresholdType in cv4? The documentation seems to be missing a description of the THRESH_MASK Thr

2019-04-14 07:10:56 -0500 commented question Failing to read pixel of grayscale image as float

Take a look at OpenCV's normalize function link text

2019-04-13 15:13:30 -0500 answered a question How to detect unclosed hexagon?

OpenCV has a useful tool called line iterator that can set up an imaginary line between 2 points. You can step along th

2019-04-13 13:32:30 -0500 commented question Failing to read pixel of grayscale image as float

As LBerger said it is hard to know what your are trying. I assume you are trying something like: Mat img(Size(256, 256)

2019-04-13 13:31:52 -0500 commented question Failing to read pixel of grayscale image as float

As LBerger said it is hard to know what your are trying. I assume you are trying something like: Mat img(Size(256, 256)

2019-04-13 13:30:58 -0500 commented question Failing to read pixel of grayscale image as float

As LBerger said it is hard to know what your are trying. I assume you are trying something like:

2019-03-13 05:19:27 -0500 commented question stretch dynamic range keeping original ratios of the color channels

Not sure exactly how you want to "adjust the dynamic range" but maybe this will give you some ideas. https://docs.opencv

2019-03-12 06:26:50 -0500 received badge  Editor (source)
2019-03-12 06:26:50 -0500 edited question OpenCV ?false? memory leak in VisualStudio

OpenCV ?false? memory leak in VisualStudio Maybe this is not a true memory leak but it must be worrisome to some develop

2019-03-12 06:24:06 -0500 asked a question OpenCV ?false? memory leak in VisualStudio

OpenCV ?false? memory leak in VisualStudio Maybe this is not a true memory leak but it must be worrisome to some develop

2019-03-11 09:41:32 -0500 commented answer Weird memory leaks in MFC

Why is this not a problem with console applications?? If I make a wxWidgets project, I also get this memory error. Howev

2019-03-04 13:20:42 -0500 commented question How to get button border size?

Invert or threshold this image, then use findContours to get the outer contour. Using this contour's center and bounding

2019-03-01 10:17:50 -0500 commented question Please suggest strategy for Part counting and identification

Many machine vision problems can benefit as much or more from improved lighting than from complex algorithms. In the or

2019-02-28 09:57:55 -0500 commented question Measuring the speed of an object with variable distance from camera

Finding the world or 3d distance between 2 points is a common issue when working with depth sensing cameras Many depth

2019-02-28 09:57:31 -0500 commented question Measuring the speed of an object with variable distance from camera

Finding the world or 3d distance between 2 points is a common issue when working with depth sensing cameras Many depth

2019-02-28 09:57:10 -0500 commented question Measuring the speed of an object with variable distance from camera

Finding the world or 3d distance between 2 points is a common issue when working with depth sensing cameras Many depth

2019-02-28 09:56:39 -0500 commented question Measuring the speed of an object with variable distance from camera

Finding the world or 3d distance between 2 points is a common issue when working with depth sensing cameras Many depth c

2019-02-28 09:36:42 -0500 answered a question How can i use opencv to get product images from market leaflet ?

If you know the product images you are looking for, template matching may work. link text

2019-02-28 09:22:34 -0500 answered a question Detection of color contrast

I assume you mean the brighter white rings on top. I was able to do this by converting the color image to HSV, then spl

2019-02-28 09:12:03 -0500 received badge  Enthusiast
2019-02-22 13:49:06 -0500 answered a question Detecting red laser color and get coordinates of the red laser color using open cv cpp code.

Have you tried splitting the image into separate r, g, b images? https://docs.opencv.org/2.4/modules/core/doc/operations

2019-02-22 11:00:17 -0500 answered a question make edges clearer in image

For many machine vision problems, lighting can be very important. I recommend playing around with the lighting to make i

2019-02-22 09:31:32 -0500 commented answer Good to see that mlp has relu activation function but returns nan??

Thanks to LBerger, I was able to get this classic xor problem to work with mlp and relu (leaky relu). The network has 2

2019-02-22 05:53:40 -0500 commented answer unsigned short 1D array of gray image

You can change from vector to array with something like this. Notice the lena image in the opencv samples is 512 x 512

2019-02-21 16:18:58 -0500 answered a question Detecting bright spots and locating the center of the bright spot

Several steps - convert the image to grayscale - threshold this mono image using some application specific threshold val

2019-02-21 14:56:58 -0500 commented answer Good to see that mlp has relu activation function but returns nan??

Yes, this works but I would like to understand why. My understanding is that relu is just max(0, d) where d is dot produ

2019-02-21 14:24:12 -0500 answered a question unsigned short 1D array of gray image

Maybe use stl::vector instead of array since different images will be different sizes. Also, data in grayscale images ar

2019-02-21 14:09:43 -0500 asked a question Good to see that mlp has relu activation function but returns nan??

Good to see that mlp has relu activation function but returns nan?? The following program uses ann_mlp to do the xor fun