Processing Forum

Motion Detection to Change Movie Speed in GSVideo

[1 Reply]
- 09-Aug-2012 10:30 AM
- Forum: Contributed Library Questions
Hi,

I'm doing a quick prototyping to check if Processing can do a motion detection via frame differencing and use the changing value to determine a movie playback speed. I stitched the FrameDifferencing and Speed examples from GSVideo example in Processing 2.0a8. The code is as follows:
2. /**
3. * GSVideo movie speed example.
4. *
5. * Use the Movie.speed() method to change
6. * the playback speed.
7. *
8. */
10. import processing.video.*;
12. Movie movie;
14. int numPixels;
15. int[] previousFrame;
16. Capture video;
18. int movementSum;
19. int newMovementSum;
21. void setup() {
22. size(320, 240);
23. background(0);
24. movie = new Movie(this, "balloon.ogg");
25. movie.loop();
27. PFont font = loadFont("DejaVuSans-24.vlw");
28. textFont(font, 24);
30. video = new Capture(this, width, height);
31. video.start();
32. numPixels = video.width * video.height;
33. // Create an array to store the previously captured frame
34. previousFrame = new int[numPixels];
35. loadPixels();
37. newMovementSum = 1;
38. }
40. void movieEvent(Movie movie) {
41. movie.read();
42. }
44. void draw() {
45. if (video.available()) {
46. // When using video to manipulate the screen, use video.available() and
47. // video.read() inside the draw() method so that it's safe to draw to the screen
48. video.read(); // Read the new frame from the camera
49. video.loadPixels(); // Make its pixels[] array available
51. movementSum = 0; // Amount of movement in the frame
52. for (int i = 0; i < numPixels; i++) { // For each pixel in the video frame...
53. color currColor = video.pixels[i];
54. color prevColor = previousFrame[i];
55. // Extract the red, green, and blue components from current pixel
56. int currR = (currColor >> 16) & 0xFF; // Like red(), but faster
57. int currG = (currColor >> 8) & 0xFF;
58. int currB = currColor & 0xFF;
59. // Extract red, green, and blue components from previous pixel
60. int prevR = (prevColor >> 16) & 0xFF;
61. int prevG = (prevColor >> 8) & 0xFF;
62. int prevB = prevColor & 0xFF;
63. // Compute the difference of the red, green, and blue values
64. int diffR = abs(currR - prevR);
65. int diffG = abs(currG - prevG);
66. int diffB = abs(currB - prevB);
67. // Add these differences to the running tally
68. movementSum += diffR + diffG + diffB;
69. //newMovementSum = movementSum/1000;
70. // Render the difference image to the screen
71. pixels[i] = color(diffR, diffG, diffB);
72. // The following line is much faster, but more confusing to read
73. //pixels[i] = 0xff000000 | (diffR << 16) | (diffG << 8) | diffB;
74. // Save the current color into the 'previous' buffer
75. previousFrame[i] = currColor;
76. }
77. // To prevent flicker from frames that are all black (no movement),
78. // only update the screen if the image has changed.
79. if (movementSum > 0) {
80. //updatePixels();
81. newMovementSum = movementSum/1000;
82. println(newMovementSum); // Print the total amount of movement to the console
83. }
84. }
86. image(movie, 0, 0, width, height);
88. //float newSpeed = map(mouseX, 0, width, 0.1, 2);
89. float newSpeed = newMovementSum/1000;
90. movie.speed(newSpeed);
91. fill(240, 20, 30);
92. text(nfc(newSpeed, 2) + "X", width - 80, 30);
93. }
It's probably long and inefficient, but then again I encountered 3 problems:
1. I keep on getting this error on the terminal line:

(Processing core video:1012): GStreamer-CRITICAL **: gst_element_send_event: assertion `event != NULL' failed

I have no idea what's that error about.

2. The video takes 3 seconds after reaching the end to loop again.

3. There's this black screen that keeps on showing every time I rapidly change the playback speed. This error also occurs in the Speed example.

Anybody has any solution for this? Or tell me if what I want is not feasible with Processing at this moment, so I can change to different tool.

Thank you :)

overlapping blobs with opencv

[0 Replies]
- 30-May-2012 05:05 AM
- Forum: Contributed Library Questions
hi, I'm trying to do blob tracking using multiple kinects. it runs fine. I draw images from each Kinect side by side, I applied opencv.threshold() on each, and everything's fine.

however, I figured that when I do blob detection, the blobs are overlapped on the image from the first Kinect. it seems that the blobs are drawn only in (0,0) coordinate. whereas, I want the blobs to be drawn for each images. is there any way to solve this?

here's the snippet of the code that I use
1. for (int i = 0; i < device_count; i++) {
2. assignPixels( depth_frame_[i], kinect_depth_[i]);
3. opencv[i].copy(depth_frame_[i]);
4. opencv[i].threshold(100);
5. image (opencv[i].image(), 640*i, 0); //no problem up to this point
6. Blob[] blobs = opencv[i].blobs(10, width*height/2, 100, true);
7. fill (240, 0, 0);
8. for (int k = 0; k < blobs.length; k++) {
9. beginShape();
10. for (int l = 0; l < blobs[k].points.length; l++) {
11. vertex (blobs[k].points[l].x, blobs[k].points[l].y);
12. }
13. endShape(CLOSE);
14. }
15. }
Cheers,

ellipse gradient stroke

[2 Replies]
- 25-Apr-2012 09:31 PM
- Forum: General Discussion
hi,

i'd like to make an ellipse with a gradient stroke that fades from white to pure transparent, similar to this

it's taken from a cover of The Mars Volta's latest album, Noctorniquet, and I really like that subtle circles. So I'm looking to recreate it in Processing. Anybody knows how to do it? I tried to hack the radial gradient example but so far I'm failed because those gradients were made to do a fill and not a stroke, per se.

Video camera streaming via USB + JMyron

[0 Replies]
- 23-Apr-2012 11:06 PM
- Forum: General Discussion
Hello everyone,

I hope I'm not asking in the wrong section here :D Anyway, I'm looking at how to stream or getting a feed from camera via USB. I have a Sony HDR CX-100 lying here in my room and I wonder can I use it as a video input using JMyron library? I understand that I can use that library with firewire port, however there are no firewire ports in both my camera and my Macbook. So, is this possible? Or do I need to add something like a USB capture card or some sort?

Thank you very much for any answers given.

Cheers.

Tuio Cursor to toggle button problem

[4 Replies]
- 26-Jan-2012 08:08 AM
- Forum: Contributed Library Questions
Hi Everyone,

I'm a bit ashamed that I have to ask this question, because this seems to be an easy work, however I've been having some issues with using TUIO cursor to toggle on/off button in Processing. I modified the ImageButton example to come up with grid of images as button, controlled by tuio cursor. I just want the tuio cursor to toggle the grid of buttons according to its position. Everything works, except for the fact that sometimes, the buttons just refused to react to the input given by the tuio cursor. I'm suspecting a simple logic fallacy here, but here's the code:
2. import TUIO.*;
3. TuioProcessing tuioClient;
5. ImageButtons[][] buttons = new ImageButtons[20][20];
7. void setup()
8. {
9. size(400, 600);
10. background(102, 102, 102);
12. // Define and create image button
13. PImage b = loadImage("butt-off-resize.jpg");
14. PImage d = loadImage("butt-on-resize.jpg");
15. int x = width/2 - b.width/2;
16. int y = height/2 - b.height/2;
17. int w = b.width;
18. int h = b.height;
20. for (int i = 0; i <20; i++) {
21. for (int j = 0; j <20; j++) {
22. buttons[i][j] = new ImageButtons (i*w, j*h, w, h, b, d);
23. }
24. }
25. tuioClient = new TuioProcessing(this);
26. }
28. void draw()
29. {
30. for (int i = 0; i < buttons.length; i++) {
31. for (int j = 0; j <buttons.length; j++) {
32. buttons[i][j].update();
33. buttons[i][j].display();
34. }
35. }
36. }
39. // called when a cursor is added to the scene
40. void addTuioCursor(TuioCursor tcur) {
41. println("add cursor "+tcur.getCursorID()+" ("+tcur.getSessionID()+ ") " +tcur.getX()+" "+tcur.getY());
42. }
44. // called when a cursor is moved
45. void updateTuioCursor (TuioCursor tcur) {
46. println("update cursor "+tcur.getCursorID()+" ("+tcur.getSessionID()+ ") " +tcur.getX()+" "+tcur.getY()
47. +" "+tcur.getMotionSpeed()+" "+tcur.getMotionAccel());
48. }
50. // called when a cursor is removed from the scene
51. void removeTuioCursor(TuioCursor tcur) {
52. println("remove cursor "+tcur.getCursorID());
53. }
55. // called after each message bundle
56. // representing the end of an image frame
57. void refresh(TuioTime bundleTime) {
58. redraw();
59. }
61. class Button
62. {
63. int x, y;
64. int w, h;
65. color basecolor, highlightcolor;
66. color currentcolor;
67. boolean over = false;
68. boolean pressed = false;
69. float tuioX, tuioY;
70. boolean state;
72. void tuioInput() {
73. Vector tuioCursorList = tuioClient.getTuioCursors();
74. for (int i=0; i<tuioCursorList.size(); i++) {
75. TuioCursor tcur = (TuioCursor)tuioCursorList.elementAt(i);
76. tuioX = tcur.getScreenX(width);
77. tuioY = tcur.getScreenY(height);
78. }
79. }
81. boolean overRect(int x, int y, int width, int height) {
82. if (tuioX >= x && tuioX <= x+width &&
83. tuioY >= y && tuioY <= y+height) {
84. return true;
85. }
86. else {
87. return false;
88. }
89. }
90. }
92. class ImageButtons extends Button
93. {
94. PImage base;
95. PImage down;
96. PImage currentimage;
98. ImageButtons(int ix, int iy, int iw, int ih, PImage ibase, PImage idown)
99. {
100. x = ix;
101. y = iy;
102. w = iw;
103. h = ih;
104. base = ibase;
105. down = idown;
106. currentimage = base;
107. }
109. void update()
110. {
111. tuioInput();
112. over();
113. if (over) {
114. state = !state;
115. }
116. else if (state == false) {
117. currentimage = base;
118. }
119. else if (state == true) {
120. currentimage = down;
121. }
122. }
124. void over()
125. {
126. if ( overRect(x, y, w, h) ) {
127. over = true;
128. }
129. else {
130. over = false;
131. }
132. }
134. void display()
135. {
136. image(currentimage, x, y);
137. }
138. }
Note: I didn't do anything to the callback method provided by the tuio library.
I hope anyone can lend me some help, I've been stuck here for a while :|

Thank you! :D

Problems with Open Kinect and Open CV

[2 Replies]
- 13-Mar-2011 07:34 PM
- Forum: Contributed Library Questions
Hello there everyone!

I was playing around with the Open Kinect library, and it was awesome, really makes it easy to do background substraction. However, I was trying to combine the depth information from the Kinect and pass it to OpenCV using the following code

import org.openkinect.*;

import org.openkinect.processing.*;

import hypermedia.video.*;

OpenCV opencv;

Kinect kinect;

void setup() {

  size (640, 480);

  opencv = new OpenCV(this);

  kinect = new Kinect (this);

  kinect.start();

  kinect.enableDepth(true);

}

void draw() {

  PImage depth = kinect.getDepthImage();

  image(depth, 0, 0);

  opencv.copy(depth);

  opencv.threshold(180);

  Blob[] blobs = opencv.blobs(10, width*height/2, 100, true);

  fill (255,0,0);

  for (int i = 0; i < blobs.length; i++) {

   beginShape();

   for (int j = 0 ; j < blobs[i].points.length; j++) {

   vertex (blobs [i].points[j].x, blobs[i].points[j].y);

   }

   endShape(CLOSE);

  }

}

void stop() {

  kinect.quit();

  super.stop();

}

, but I always get an error message like this

OpenCV ERROR: Null pointer (NULL array pointer is passed)

Invalid memory access of location 00000000 eip=22c384bf

in function cvGetMat, ../../src/cxcore/cxarray.cpp(2781)

Terminating the application...

I couldn't see mistakes in my code, as I've presumably followed required steps for using both library.

Can anybody point out where my mistake could be?

Thank you very much guys :)

How to count unique elements in Arraylist?

[1 Reply]
- 25-Jan-2011 01:01 AM
- Forum: Programming Questions
Hi there guys :D

Is there anyway to count unique element in arraylist? I have a number of elements inside an arraylist and I want to count the frequency of each element's appearance. I understand that there's a method such as frequency for Java, but I don't know how to do this natively in Processing.

Thanks for your help :D
Cheers

Object Detection from Webcam

[3 Replies]
- 04-Oct-2010 06:12 AM
- Forum: General Discussion
Hi There,

I'm working on a school project, in which I want to detect and track a company logo in a business card using a webcam. The question is pretty much basic, I'm kinda lost what feature I should use? I mean i've tried the face detection on open cv library, but it always ended up tracking my face. I've also tried blob detection, but I don't know how to keep on tracking that logo only, because it appears that it detects all of the shape that it founded.

Sorry if this question seems basic, because I'm pretty much lost in this computer vision field.

Thank you for any help :)

how to create a multiple line of text without using text field?

[3 Replies]
- 02-Oct-2010 08:36 PM
- Forum: Programming Questions
hi there,

does anyone know how to create a multiple line of text? I know how to create a canvas and how to start type on it, but the problem is, after the string's width is bigger than the canvas' width, I want to automatically set the typing cursor to be below the current typed word. I understand that I might be able to achieve this using text field, but I'm not that aesthetically pleased with having a text field on my canvas. So is there any solution?

Thank you very much guys :)

Activity Trend