In this example, I demonstrate the use of the image processing module (Imgproc) with the filter2D function. It is similar to the generic Photoshop filter where you can design the convolution matrix. The example references the OpenCV tutorial and uses a sharpen filter. It also uses the CVImage class in the last post.
Here is a screen shot of the Processing window.
I decide to put together the OpenCV and Processing codes into a class to encapsulate the functions. In this example, I extend the original PImage class and create the CVImage class. Besides the constructor, the following functions are the major interfaces:
Similar to the previous post, it is the Java build of the official release of OpenCV 3.0.0 with the 64 bit versions of Linux, Mac OSX, and Windows. It is now available for download here at
OpenCV 3.0.0 Java build (64 bit)
Copy them to the code folder of any Processing sketch to play.
The example recreates the VideoCapture sample from the OpenCV documentation. It uses the Imgproc module to convert the color image into single channel greyscale, apply a blur filter, and find the edges.
The example converts a PImage (ARGB) to a Mat (BGR). It then randomly samples a pixel to compare their color values in the two formats. In order to transfer the integer array pixels[] from the PImage to the byte array of the CV_8UC4 matrix, it makes use of the ByteBuffer and IntBuffer.
Continue reading
This example compares the pixel information between a Processing PImage and OpenCV Mat of the same external image file. The dimension of the image in Processing is width x height (x, y) while for OpenCV, it is rows() x cols() (y, x). A random point is picked and its pixel color is shown both in Processing PImage and OpenCV Mat.
Continue reading
To perform the same image loading task in OpenCV, we use the imread() function in the Imgcodecs module. The function was in the Highgui module before. The image loaded in the Mat data structure is in BGR format. We shall use the split and merge commands to align the proper channels. Finally, we use a byte array to transfer the pixels information to the pixels[] array of a PImage object.
Continue reading
Now we proceed to first core part of the tutorials, representation of image in Processing and OpenCV. In Processing, the class is PImage. Digital video, Movie and live webcam feed, Capture are also PImage. When we import the external image file from the data folder through loadImage(), the format will usually be RGB. The internal representation is, however, ARGB.
Continue reading
This example will demonstrate the use of the Rect class in OpenCV and compare it with the Rectangle class in Java. The Rect class will be useful when we work on the face detection example later. The detected faces will be returned as rectangles.
Continue reading
