Robotics@Cyborg

Go-visible

2012-02-09T06:16:00.000-08:00

Finally got my kinect to work, thanks to the CL NUI library :)
I am now able to access the image as well as the depth map from the kinect, in rgb format.

This is the first thing that I am trying with my kinect. What I have done is thresholding the depth map to find objects within a particular image.

Virtual keyboard

2011-12-03T08:48:00.000-08:00

Today virtual projection keyboards are a craze. It is used as modules which can be connected to computers like the cellulon projection keyboard, are used in mobile, PDA's, and small equipments, where space is very small forcomputing.

I, myself am dying to have one. So, inspired by it, I take it as my seminar topic. I was much interested in making one myself also, but I didnt have either the red laser diode projector or the IR laser diode for the detection plane. So, I made a short program in OPENCV to immitate it, by just taking 11 keys.

The text being typed can be seen at the bottom of the video.

I have used color marker on fingers along with blob detection to get these results.

My new Kinect :)

2011-12-03T06:32:00.000-08:00

Hiii friends.
Finally I was able to get myself a kinect.
Kinect is a motion sensing input device developed by Microsoft for the Xbox 360 gaming console. Based around a webcam-style add-on peripheral for the Xbox 360 console, it enables users to control and interact with the Xbox 360 without the need to touch a game controller, through a natural user interface using gestures and spoken commands.
Now, it is also being used with P.Cs for developing applications for kinect for easier Human computer interaction systems.
There's a trio of hardware innovations working together within the Kinect sensor:

Color VGA video camera - This video camera aids in facial recognition and other detection features by detecting three color components: red, green and blue. Microsoft calls this an "RGB camera" referring to the color components it detects.
Depth sensor - An infrared projector and a monochrome CMOS (complimentary metal-oxide semiconductor) sensor work together to "see" the room in 3-D regardless of the lighting conditions.
Multi-array microphone - This is an array of four microphones that can isolate the voices of the players from the noise in the room. This allows the player to be a few feet away from the microphone and still use voice controls.

This is the snapshot of my kinect :)

While I was trying to figure out how to use it, I came across CL NUI platform for kinect users, at http://codelaboratories.com/downloads/ .

It was just cool. There is no command line required to use it. All u have to do is plug in your kinect, install it and run it. It gives you the functionality to see the image of camera, the depth data. Moreover, it also has the feature to move the kinect to face up and down to some angle using motors, which was really cool.

U.G.V(Unmanned Ground Vehicle)

2011-07-29T13:14:00.000-07:00

It started with the "DRDO DRIVING INNOVATION STUDENT ROBOTICS COMPETITION 2010". This is where everything started, it has been one hell of a time since then.

The basic requirements for making a UGV are:

Camera - to take visual input and find the path.
Laser range detector - for detecting obstacles and avoiding them.
Locomotion system or the chasis along with motors and wheels for making the bot move.
GPS- for tracking its position.
Wireless module for E-Stop.
A team with madly interested people. In my case, our team consisted of my friends Deepankar, Gigyanshu, Rahul, Keshav and Abhishek.
A lot of night outs and workshop deliveries :)

We started off with Matlab. But as it was very slow, we switched on to OpenCV.
This is the video of our UGV following the lane.

An example of lane detection by our UGV can be seen below:

In the video, the bot is working only through lane detection and there is Wireless E-Stop for emergency.

Now, we have added the obstacle detection and avoidance to it, as well as GPS tracking.

Next time I will be posting the videos regaurding obstacle avoidance, and other modifiactions to it..

Untill then....Next time.....

Augmented reality

2011-07-29T08:48:00.000-07:00

I have been trying to do augmented reality since a long time, and while surfing, I came across the ARToolkit and SUDARA.

How to make your own haar trained ".xml" files

2011-07-29T08:29:00.000-07:00

hiii friends
Today i write on how to make ur own xml file. It took me long to figure it out.
The basic requirement is for u to have OpenCV installed and integrated to ur IDE.
If u haven't, then u could do it by following this link:
http://nayakamitarup.blogspot.com/2011/05/introduction-to-opencv-for-image.html

Now that u have installed OpenCV, u r ready to go.

To start, dowload the "tools.exe" from the link below:
http://www.4shared.com/file/mpbQkWp4/tools.html

Unzip the tools.exe into a folder. U will find the contents like given below:

For training, u would need a database of thousands of images, with and without the object u want to train the xml file for, respectively called as positive and negative images. Near about 1000-2000 positives and 3000 negetives, sound fine for training.

Now u need to place all the positive images(images with object) inside the folder "\temp\positive\rawdata".
Now that u have placed all the positives, execute the "objectmarker.exe". On executing it, it will show the image, and u need to mark the area containing the object. and u get some thing like this:

Then press "Space Bar". It will feed the coordinates of the bounding rectangle into a text file "info.txt".
Then Press "Enter" to go to next image, and continue similarly till u are done with all the images.
This is the most tiring part, and makes u wanna kill urself :P .

WARNING: Donot kill urself, have patience. :)

Now u can see the "info .txt" looks somthing like this:

Phew!!!!
Done with the most tiring part. Now insert all the negetive images(images without the object) or the so called background images into the folder "temp\negative", and execute the "create_list.bat" file.
On executing it, it stores the names of negative images along with extension, into a text file "infofile.txt" and it looks something like this:

Now u need to create samples for haar training. It is done by typing the following command in "start->run"

C:\Users\amit\Desktop\hand\lasssssssssssssst\temp\createsamples.exe -info positive/info.txt -vec data/vector.vec -num 527 -w 24 -h 24

Here, to use this command u need to edit it , with the number of positive and negetive images. Since I had used 527 positive images and 1142 negetive images and 30 stages, u can change it according to use by simply replacing the numbers in the command.

Now that the samples have been created, u can start haar training.
The same procedure is to be followed. Type the command in "start->run".

C:\Users\amit\Desktop\hand\lasssssssssssssst\temp\haartraining.exe -data data/cascade -vec data/vector.vec -bg negative/infofile.txt -npos 527 -nneg 1142 -nstages 30 -mem 1000 -mode ALL -w 24 -h 24 -nonsym

This starts the haar training, and u can see on the window, something like this:

This will continue upto the stage u have mentioned. It takes hell lot of a time, and during this time it occupies nearly 60% of the processor.
As the training proceeds,u will find folders forming with the text file "AdaBoostCARTHaarClassifier.txt" forming and a "vector.vec" file is also formed.

On completion of training, replace the "data" in "haar\cascade2xml" by the "data" folder in "haar\temp", and copy the "vector.vec" to location "haar\cascade2xml".
Now execute the "convert.bat" file. It finally forms the output.xml file.

Congrats, u made ur own xml file :)

My summer project :)

2011-07-08T08:59:00.000-07:00

Hii
Today I and Gigyanshu feel greatly relieved having completed a part of our summer intern project. Our project deals with Image processing using OpenCv to detect shape and color of objects and Programming the " SPARKV " robot to find them.

We have used my faithful ACER 5920 laptop (2.00 GB RAM and 2.20 GHz with Intel Core 2 Duo), a 16 MP webcam, 2 Xbee wireless modules.

SparkV Bot

This is the video of result until now. Still need to improve it, so we would be posting more videos with better results :)

2D AR using "SURF" based application of OpenCV

2011-06-29T12:56:00.000-07:00

This is 2D AR using "SURF" based application of OpenCV. SURF (Speeded Up Robust Features) is a robust image detector & descriptor.

OpenCV: Alien Attack

2011-06-16T00:08:00.000-07:00

Aliens have attacked and u r the sole survivor. So, to survive, u need to slap them to death. :)
I have been playing it for an hour nearly and am loving it. :)

It has been made using OpenCV and C++. It is an addition to the interface in the previous post. Hope u like it.

To enjoy it U have to download the file games.rar from the link below
http://www.4shared.com/get/9H5lh6vs/game.html
Unzip it in ur E drive (important) directly. Run the game.exe to execute. Njoy :)

Application using OpenCV to play music, draw, play video using hand gestures

2011-06-09T09:10:00.000-07:00

Hii friends, this is something me n my friend Prakash came up with in order to kill the boredum of our summer holidays :) .

This is made in OpenCV using Code::Blocks IDE. We have used gesture recognition as the prime component.

This is incomplete and we will be adding more features like games, giving urself wierd hairstyles using face detection, doing matematics using gestures and much more features.

Gesture recognition using Haar clasifier

2011-06-02T08:56:00.000-07:00

Hiii, friends since some days, I have been working on Gesture recognition using using Haar-like features. I havent been able to get very good results, owing to the lack of hand database. More over, the Classifier is taking hell lot of a time to get trained.

This is the result I got by using haar clasifier to train it for 2 gestures, my fist xml file being of nearly 70 kb, and xml file for L palm being nearly 30 kb. They are indeed small, and so the result i get , u can see from the video.

For better results, I need to train my classifier with more positive images, nearly 3000, and negetive images nearly about 5000. Now, I am making my own collection of hand gestures :)
.
I heard of a Massey's Gesture database, but the link doesnt seem to be working. ( sighs )

But still u can use this database
http://www.iis.ee.ic.ac.uk/~tkkim/ges_db.htm

I have started my training my new .xml files for closed fist or 'A' using 527 positive and 1142 negetive images, and it is taking like forever.( sighs ). I guess it may get fully trained within about three days. Then I will be posting my new results :)

Using the results I get above, I tried simulating mouse movements, and clicking, I used my closed fist gesture to move the mouse, and the L palm for clicking. It wasnt that stable, but still, it works :) (cheers). Its fun..

Swarm Robotics@ IIT Bombay Techfest 2010.

2011-05-27T10:18:00.000-07:00

Me and my friends took part in a very interesting robotics event ANTZ, based on swarm intelligence, conducted by IIT Bombay during Techfest 2010. It had Indian participants as well as International participants from Thailand, Sri Lanka and Australia.The initial rounds were conducted nationally to select 3 top teams each from India, Austalia, Sri Lanka and Thailand. Then finally the 12 teams contested at the finals.

Our team was the winner(1st) of ANTZ WORLD FINALE. :)

Our team.From left-Vikas, Deepankar Singh, Rahul Sinha and me.

http://mondaymorning.nitrkl.ac.in/index.php/campus/campus-buzz/6-nit-rourkela-team-wins-international-robotics

Swarm robotics basically includes a number of bots communicating with each other to coordinate an action.
Our event required 2 bots to communicate with each other to pick up boxes placed randomly in the arena and place the blocks in the certain required area. There were also included regions that were 1 bot zone, i.e. only 1 bot could enter at a time and similarly many more rules need to kept in mind.

Some snaps of the bots which took part in the event:

The video of our bot during the initial practice round. Our bot running on 100 rpm motors, in slow poke mode. lol!!

The only problem with this was speed. Our algorithm was perfect and we never got any fault, but the problem was speed. So, we doubled its speed by using 200 rpm motors and changed calibration, and now it was just rock n rolla!! :)
Sadly our camera was lost on the last day ( sighs ), so can't upload our final run video. :(

Still njoy the videos of other bots:

Now lets talk about swarm robotics. The main component of concern here is the communication between bots. For this purpose, these days a lot of modules are available, but the modules which are mainly used for hobby robotics are:

1) CC2500 2.4Ghz RF transceiver module

2) Zigbee module

Using the above modules we communicate regaurding the position and status of bot, and thus plan our actions accordingly.

P.O.V. (Persistence Of Vision)

2011-05-26T11:07:00.000-07:00

This is the video of POV made by me n my friends Deepankar Singh, Rahul Sinha.
If this video doesnt work, then see the link below
http://www.youtube.com/watch?v=J00ZKz__UKc

To make it we used the components:
1)Microcontroller - AtMega32
2)Leds - 7.(u can use any number according to display u desire)
3) LM 7805 - 5v supply
4)TSOP - for sensing IR blinking at 38khz, to be used as external interrupt
5)Capacitors-100 microfarad

The circuit diagram is:

The whole circuit is stuck to a fan. Then we give an external IR source blinking at 38khz at constant point, so that when the circuit rotates, it meets the ir light at a constant point always.This triggers low output in the TSOP. The TSOP is generally high in absence of 38khz IR, but goes low in the presence of IR.
This can be used to detect when it has completed an entire circle of revolution. For this purpose, we can use interrupts or simple poling i.e. checking the pin continiously.
Then the time taken by it is stored and used for displaying letters. We divide the time into total no. of sectors in which we want to display.

We have used 7x5 column to display letters, i.e. we have used 5 sectors of led positions, manipulating their on and off state, to display. We can use the BITMAP CALCULATOR which lets you draw a character and
visualise it. It can be seen in
http://www.geocities.com/dinceraydin/lcd/charcalc.htm

Now, when the circuit starts rotating at a nearly constant speed, the first ir signal received indicates the start of rotation, the next signals received indicate the end of rotation and hence the rotation time value is stored and then the display starts. For display, lets suppose we want to display C, then, it must be like below

Seeing this we can say that the sequence of glowing of leds must be
11111111
then
10000001
then
10000001
then
10000001
then
10000001
and last
11100111

This completes the display of "C". The wideness of display depends on the time alloted for the display of each line. Similarly we can diplay other characters also.

Handling mouse events

2011-05-14T07:06:00.000-07:00

We can monitor the mouse movements to trigger different events, lets start with setting a ROI( Region Of Interest ) using the clicking of left mouse button.

The code is :

#include "cv.h"
#include "highgui.h"
CvPoint point;
int drag = 0;
void mouse_callback(int event, int x, int y, int flags, void* img)
{
if (event == CV_EVENT_LBUTTONDOWN)
{
    point = cvPoint(x, y);
    drag = 1;
}
else if (event == CV_EVENT_MOUSEMOVE && drag)
{
    IplImage* img1 = cvCloneImage((IplImage*) img);
    cvRectangle(img1, point, cvPoint(x, y), CV_RGB(255, 0, 0), 1, 8, 0);
    cvShowImage("img", img1);
}
else if (event == CV_EVENT_LBUTTONUP)
{
    drag = 0;
}
}

int main()
{
IplImage* img = cvLoadImage("C:/Users/amit/Desktop/myblog/ball detection/pan.jpg");

cvNamedWindow("img", 1);

cvSetMouseCallback("img", mouse_callback, img);
cvShowImage("img", img);

cvWaitKey(0);

cvDestroyAllWindows();
cvReleaseImage(&img);
return 0;
}

Explanation
What happens in this code is that when we click the left mouse, then the event is set to CV_EVENT_LBUTTONDOWN, which implies the pressing of left mouse button and the drag is set. Now on dragging the mouse with left button pressed, the rectangle is formed with corners, one end with where it was first clicked and where it is now.
When we release the left button, then the event is set to CV_EVENT_LBUTTONUP, and thus we get the ROI as required.

Application:

1. It has a very good application. We can use it to track an object of a color let it be a ball like we described in previous tutorial. What we do is we select the ROI and thus find the ranges of HUE, SATURATION and VALUE for the ball color, and then using shape detection like before, we can detect ball.

The code is :

#include "cv.h"
#include "highgui.h"
#include <iostream>
using namespace std;
CvPoint point;
int drag = 0;
int hmin=180,hmax=0,smin=256,smax=0,vmin=256,vmax=0,cnt=0,xs,ys,xe,ye;
IplImage* frame = NULL;
void mouse_callback(int event, int x, int y, int flags, void* img)
{
if (event == CV_EVENT_LBUTTONDOWN)
{
    point = cvPoint(x, y);
    drag = 1;cnt=10;
    if(cnt==10)
    {xs=x;ys=y;}
}
else if (event == CV_EVENT_MOUSEMOVE && drag)
{
    IplImage* img1 = cvCloneImage((IplImage*) img);
    //cvRectangle(img1, point, cvPoint(x, y), CV_RGB(255, 0, 0), 1, 8, 0);
    //cvShowImage("img", img1);
    cvRectangle(frame, point, cvPoint(x, y), CV_RGB(255, 0, 0), 1, 8, 0);
    cvShowImage("video", frame);
    if(cnt==10)
    {xe=x;ye=y;}
}
else if (event == CV_EVENT_LBUTTONUP)
{
    drag = 0;
    if(cnt==10)
    {cnt=cnt+1;}
}

}
main()
{
int i,j,xa,ya;
     CvCapture* capture = cvCaptureFromCAM( 0 );
     if( !capture )
     {
             fprintf( stderr, "ERROR: capture is NULL \n" );
             getchar();
             return -1;
     }
    cvNamedWindow("video", CV_WINDOW_AUTOSIZE);
    cvNamedWindow("colordetect", CV_WINDOW_AUTOSIZE);

frame = cvQueryFrame( capture );
    CvSize size = cvGetSize(frame);
IplImage * stor    = cvCreateImage(size, IPL_DEPTH_8U, 3);
     IplImage * hsv_frame    = cvCreateImage(size, IPL_DEPTH_8U, 3);
     IplImage* thresholded   = cvCreateImage(size, IPL_DEPTH_8U, 1);
// IplImage* img = cvLoadImage("C:/Users/amit/Desktop/myblog/ball detection/pan.jpg");
//cvNamedWindow("img", 1);
while(1)
     {

        //capture image from cam
         frame = cvQueryFrame( capture );
        // stor=cvCloneImage(frame);
         if( !frame )
         {
                 fprintf( stderr, "ERROR: frame is null...\n" );
                 getchar();
                 break;
         }

cvSetMouseCallback("video", mouse_callback, frame);
cvShowImage("video", frame);
//cvShowImage("colordetect", thresholded);
if(cnt==11)
{cnt=0;
cvDestroyAllWindows();

cvNamedWindow("video", CV_WINDOW_AUTOSIZE);
    cvNamedWindow("colordetect", CV_WINDOW_AUTOSIZE);
    cvCvtColor(frame, hsv_frame, CV_BGR2HSV);

for(xa=xs+1;xa<xe;xa++)
{
for(ya=ys+1;ya<ye;ya++)
{
CvScalar s=cvGet2D(hsv_frame,xa,ya);
if(hmin > s.val[0])
{hmin = s.val[0];}
if(hmax < s.val[0])
{hmax = s.val[0];}

if(smin > s.val[1])
{smin = s.val[1];}

if(smax < s.val[1])
{smax = s.val[1];}
if(vmin > s.val[2])
{vmin = s.val[2];}

if(vmax < s.val[2])
{vmax = s.val[2];}
cout<<s.val[0]<<endl;

}
}
cout<<"thresholded"<<endl;
cout<<hmax<<" "<<hmin<<endl;
cout<<smax<<" "<<smin<<endl;
cout<<vmax<<" "<<vmin<<endl;

CvScalar hsv_min = cvScalar(hmin, smin, vmin, 0);
    CvScalar hsv_max = cvScalar(hmax, smax, vmax, 0);
    while(1)
    {
        frame = cvQueryFrame( capture );
        if( !frame )
         {
                 fprintf( stderr, "ERROR: frame is null...\n" );
                 getchar();
                 break;
         }
         cvCvtColor(frame, hsv_frame, CV_BGR2HSV);
        //color detection
        cvInRangeS(hsv_frame, hsv_min, hsv_max, thresholded);
         // Memory for hough circles
        // CvMemStorage* storage = cvCreateMemStorage(0);
        cvSmooth( thresholded, thresholded, CV_GAUSSIAN, 11, 11 );
CvMemStorage* storage = cvCreateMemStorage(0);
        //shape detection-circle shape
         CvSeq* circles = cvHoughCircles(thresholded, storage, CV_HOUGH_GRADIENT, 2,thresholded->height/6, 100, 50, 10, 400);
         cout<<"no. of circles="<<circles->total;
         //drawing circle around the detected circle
         for (int i = 0; i < circles->total; i++)
         {
             float* p = (float*)cvGetSeqElem( circles, i );
             cout<<"x= "<<p[0]<<"y= "<<p[1]<<"r= "<<p[2]<<endl;
             cvCircle( frame, cvPoint(cvRound(p[0]),cvRound(p[1])),cvRound(p[2]), CV_RGB(255,0,0), 3, 8, 0 );
         }

cvShowImage("video", frame);
cvShowImage("colordetect", thresholded);
cvReleaseMemStorage(&storage);
        if( (cvWaitKey(10) & 255) == 27 ) { //hmin=180,hmax=0,smin=256,smax=0,vmin=256,vmax=0;
            break;}
        }

    }

if( (cvWaitKey(10) & 255) == 27 ) break;
     }
cvReleaseCapture( &capture );
cvDestroyAllWindows();
//cvReleaseImage(&);
}

2. U can make ur Pong game in OpenCV. U can check out this video in youtube:
http://www.youtube.com/watch?v=abusTzgsYYM

Detection of shapes like quads,rectangle,triangle and other complex shapes using OpenCV

2011-05-14T05:47:00.000-07:00

To detect the shapes we would be utilising the property contour. It can be thought of more like the number of corners of a quadilateral in an image. Based on the number of contour and the angle correlation we can even detect complex shapes.

The function of most importance here is cvFindContours(). Its format is:

cvFindContours(IntPtr image,IntPtr storage,IntPtr% firstContour,int headerSize,RETR_TYPE mode,CHAIN_APPROX_METHOD method,Point offset)

Parameters

image (IntPtr): The source 8-bit single channel image. Non-zero pixels are treated as 1s, zero pixels remain 0s - that is image treated as binary. To get such a binary image from grayscale, one may use cvThreshold, cvAdaptiveThreshold or cvCanny. The function modifies the source image content

storage (IntPtr): Container of the retrieved contours

firstContour ( IntPtr %): Output parameter, will contain the pointer to the first outer contour

headerSize (Int32): Size of the sequence header, >=sizeof(CvChain) if method=CV_CHAIN_CODE, and >=sizeof(CvContour) otherwise

mode (RETR_TYPE): Retrieval mode

method (CHAIN_APPROX_METHOD): Approximation method (for all the modes, except CV_RETR_RUNS, which uses built-in approximation).

offset (Point): Offset, by which every contour point is shifted. This is useful if the contours are extracted from the image ROI and then they should be analyzed in the whole image context

Now lets check it on an image. Let the input image after thresholding be

STEP1: Convert image to gray scale image

cvCvtColor(img, grayimg, CV_BGR2GRAY);

This converts the BGR image to Gray scale image and store as grayimg.

STEP2: Find contours

cvFindContours(IntPtr image,IntPtr storage,IntPtr% firstContour,int headerSize,RETR_TYPE mode,CHAIN_APPROX_METHOD method,Point offset

It finds the contours, their coordinates and stores it .Then on analyzing it and the angles,we can infer the shapes.

STEP3: Toggle through the contours and find the number of points in a contour

We then toggle through the contours to find the number of points in a contour is 4, it can be infered as quadilateral, and if the angles are nearly 90 degree then we can label it as rectangle. If the number of points in a contour is 3, then it is a triangle and so on we can detect other complex shapes also.

//detecting quad with 4 sides

while(contours)
    {
        result = cvApproxPoly(contours, sizeof(CvContour), storage, CV_POLY_APPROX_DP, cvContourPerimeter(contours)*0.02, 0);

        if(result->total==4 && fabs(cvContourArea(result, CV_WHOLE_SEQ)) > 20)
        {
            CvPoint *pt[4];

            for(int i=0;i<4;i++)
                pt[i] = (CvPoint*)cvGetSeqElem(result, i);

            cvLine(ret, *pt[0], *pt[1], cvScalar(255));
            cvLine(ret, *pt[1], *pt[2], cvScalar(255));
            cvLine(ret, *pt[2], *pt[3], cvScalar(255));
            cvLine(ret, *pt[3], *pt[0], cvScalar(255));

            cout<<pt[0]->x<<" "<<pt[0]->y<<" "<<angle(pt[0],pt[2],pt[1])<<endl;
            cout<<pt[1]->x<<" "<<pt[1]->y<<" "<<angle(pt[1],pt[3],pt[2])<<endl;
            cout<<pt[2]->x<<" "<<pt[2]->y<<" "<<angle(pt[2],pt[0],pt[3])<<endl;
            cout<<pt[3]->x<<" "<<pt[3]->y<<" "<<angle(pt[3],pt[1],pt[0])<<endl;
            cout<<endl;
n++;
xt[n]=(int)((pt[0]->x + pt[1]->x + pt[2]->x + pt[3]->x)/4);
yt[n]=(int)((pt[0]->y + pt[1]->y + pt[2]->y + pt[3]->y)/4);
}
contours = contours->h_next;
        }

STEP4: Use text to write the shape of a quad at its centroid

In the previous tutorial u have been through text adding.
Use it to add the text regaurding shape at the centre of figure.

//initiate font and font size
cvInitFont(&font, CV_FONT_HERSHEY_SIMPLEX, 1.0, 1.0, 0, 1, CV_AA);

//write text in red color
cvPutText(contourDrawn, "Rectangle", cvPoint(x,y), &font, cvScalar(255, 0, 0, 0));

Now u r ready for the final prog. :)

Do tell me if u make something new out of it. ^_^

//code to detect rectangle and rectangle.U can modify it for other shapes

#ifdef _CH_
#pragma package <opencv>
#endif
#define CV_NO_BACKWARD_COMPATIBILITY
#include "cv.h"
#include "highgui.h"
#include <stdio.h>
#include <math.h>
#include <string.h>
#include <iostream>
using namespace std;

CvFont font;
int xt[20],yt[20],n=-1,m=-1,xtr[20],ytr[20];

double angle( CvPoint* pt1, CvPoint* pt2, CvPoint* pt0 )
{
    double dx1 = pt1->x - pt0->x;
    double dy1 = pt1->y - pt0->y;
    double dx2 = pt2->x - pt0->x;
    double dy2 = pt2->y - pt0->y;
    return (dx1*dx2 + dy1*dy2)/sqrt((dx1*dx1 + dy1*dy1)*(dx2*dx2 + dy2*dy2) + 1e-10);
}

IplImage* DetectAndDrawQuads(IplImage* img)
{
    CvSeq* contours;
    CvSeq* result;

    CvMemStorage *storage = cvCreateMemStorage(0);

    IplImage* ret = cvCreateImage(cvGetSize(img), 8, 3);
    IplImage* temp = cvCreateImage(cvGetSize(img), 8, 1);

    cvCvtColor(img, temp, CV_BGR2GRAY);

    cvFindContours(temp, storage, &contours, sizeof(CvContour), CV_RETR_LIST, CV_CHAIN_APPROX_SIMPLE, cvPoint(0,0));

    while(contours)
    {
        result = cvApproxPoly(contours, sizeof(CvContour), storage, CV_POLY_APPROX_DP, cvContourPerimeter(contours)*0.02, 0);

        if(result->total==4 && fabs(cvContourArea(result, CV_WHOLE_SEQ)) > 20)
        {
            CvPoint *pt[4];
            for(int i=0;i<4;i++)
                pt[i] = (CvPoint*)cvGetSeqElem(result, i);

//to eliminate the outer recatngle,replcae 1 & 398 by
// the respective min and max value of x
if(pt[0]->x==1 && pt[1]->x==1 && pt[2]->x==398 && pt[3]->x==398)
{}

else
{
           cvLine(ret, *pt[0], *pt[1], cvScalar(255));
            cvLine(ret, *pt[1], *pt[2], cvScalar(255));
            cvLine(ret, *pt[2], *pt[3], cvScalar(255));
            cvLine(ret, *pt[3], *pt[0], cvScalar(255));

            cout<<pt[0]->x<<" "<<pt[0]->y<<" "<<angle(pt[0],pt[2],pt[1])<<endl;
            cout<<pt[1]->x<<" "<<pt[1]->y<<" "<<angle(pt[1],pt[3],pt[2])<<endl;
            cout<<pt[2]->x<<" "<<pt[2]->y<<" "<<angle(pt[2],pt[0],pt[3])<<endl;
            cout<<pt[3]->x<<" "<<pt[3]->y<<" "<<angle(pt[3],pt[1],pt[0])<<endl;
            cout<<endl;

n++;
xt[n]=(int)((pt[0]->x + pt[1]->x + pt[2]->x + pt[3]->x)/4);
yt[n]=(int)((pt[0]->y + pt[1]->y + pt[2]->y + pt[3]->y)/4);
}
        }
    else if(result->total==3 && fabs(cvContourArea(result, CV_WHOLE_SEQ)) > 20)
        {
            CvPoint *pt1[3];

            for(int i=0;i<3;i++)
                pt1[i] = (CvPoint*)cvGetSeqElem(result, i);
            cvLine(ret, *pt1[0], *pt1[1], cvScalar(255));
            cvLine(ret, *pt1[1], *pt1[2], cvScalar(255));
            cvLine(ret, *pt1[2], *pt1[0], cvScalar(255));

            cout<<pt1[0]->x<<" "<<pt1[0]->y<<" "<<angle(pt1[0],pt1[2],pt1[1])<<endl;
           ;
            cout<<endl;

m++;
xtr[m]=(int)((pt1[0]->x + pt1[1]->x + pt1[2]->x)/3);
ytr[m]=(int)((pt1[0]->y + pt1[1]->y + pt1[2]->y)/3);
        }
        contours = contours->h_next;
    }
    cvReleaseImage(&temp);
    cvReleaseMemStorage(&storage);
    return ret;
}

int main()
{
    int i;
   IplImage* img = cvLoadImage("C:/Users/amit/Desktop/iitb/1.png");

   IplImage* contourDrawn = 0;
   cvNamedWindow("original");

   cvInitFont(&font, CV_FONT_HERSHEY_SIMPLEX, 1.0, 1.0, 0, 1, CV_AA);
    contourDrawn = DetectAndDrawQuads(img);

    if(n!=-1)
    {
    for(i=0;i<=n;i++)
    {
    cvPutText(contourDrawn, "Rectangle", cvPoint(xt[i],yt[i]), &font, cvScalar(255, 0, 0, 0));
    }
    }

    if(m!=-1)
    {
    for(i=0;i<=m;i++)
    {
    cvPutText(contourDrawn, "triangle", cvPoint(xtr[i],ytr[i]), &font, cvScalar(255, 0, 0, 0));
    }
    }

   cvShowImage("original", img);
    cvNamedWindow("contours");

    cvShowImage("contours", contourDrawn);
    cvWaitKey(0);
    return 0;
}

DTMF(Dual Tone Multi Frequency) or Mobile controlled bot

2011-05-14T01:33:00.000-07:00

DTMF is Dual Tone Multi Frequency.The circuit is given by

The components required are:

1) DTMF Decoder IC - MT8870 or CM8870
2) atmega32
3) l293d
4) cd7004
5) 100k resistances – 2 no.
6) 330k resistances
7) 10k resistances-2 no
8) 0.47 micro farad capacitor
9) 22pfarat capacitor – 4 no
10) 0.1micro farad capacitor
11) 3.57 mhz crytal
12) 12mhz crystal
13) push to on switch
14) 12v 100 rpm motor – 2no.

15) Battery - 6v
16) 74LS04-not gate
17) IN4007 diode
18) LM7805

Conventionally, Wireless-controlled robots use RF circuits, which have the drawbacks of limited working range, limited frequency range and the limited control. Use of a mobile phone for robotic control can overcome these limitations. It provides the advantage of robust control, working range as large as the coverage area of the service provider, no interference with other controllers and up to twelve controls.

For signaling, it will use dual-tone multi-frequency (DTMF) tones. These are the tones that you hear when you press keys on a telephone keypad. The CM8870 IC or MT8870 IC interprets the tones as a 4-bit digital signal, which is input to microcontroller Atmega32.

So then based on which key is pressed, the robot will be able to interpret the tone and activate relays and such to turn on motors and move about.

We use Handsfree headset of phone, which consists of TIP,which is connected to mt 8870, and ring and ground are grounded.

The picture of my bot is given below. I know it looks rough, but this is what I along with my friends Sunil and Virupaksh could manage within one night :)

The code is

#include<avr/io.h>
#include<util/delay.h>
#include<lcd4.c>
#define CHECKBIT(x,b) x&(1<<b) //Checks bit status
#define SETBIT(x,b) x|=(1<<b) //Sets the particular bit
#define CLEARBIT(x,b) x&=~(1<<b) //Sets the particular bit
#define TOGGLEBIT(x,b) x^=(1<<b) //Toggles the particular bit

int check1()
{

if((PINA & 0b00000001)==0b00000001)
return 1;

else
return 0;

}

int check2()
{

if((PINA & 0b00000010)==0b00000010)
return 1;

else
return 0;

}

int check3()
{

if((PINA & 0b00000100)==0b00000100)
return 1;

else
return 0;

}

int check4()
{

if((PINA & 0b00001000)==0b00001000)
return 1;

else
return 0;

}

int main()
{
DDRA=0x00;DDRD=0xFF;
unsigned int k, h;
port();
initiate();
pos(1,1);string("techieNITRKL",0);
int a=0,b=0,c=0,d=0,f=0;

//PORTA=0x00;
while(1)
{

a=check1();
b=check2();
c=check3();
d=check4();
f=((a*1000) + (b*100) + (c*10) + d);

pos(2,1);number(f);

if(f==1110)
{
pos(2,7);number(2);//forward
PORTD=0b00000101;
}

else if(f==1011)
{
pos(2,7);number(4);//left
PORTD=0b00000111;
}

else if(f==1001)
{
pos(2,7);number(5);//back
PORTD=0b00001010;
}

else if(f==1010)
{
pos(2,7);number(6);//right
PORTD=0b00001101;
}

else
{
pos(2,7);string("P",0);
PORTD=0b00001111;
}

}

return 0;
}

Add text to image

2011-05-13T21:24:00.000-07:00

The code is:

#include "cv.h"

#include "highgui.h"

#include <iostream>

using namespace std;

int main()

{

IplImage *img = cvLoadImage("C:/Users/amit/Desktop/BACKUP1.jpg");

/* initialize font and add text */

CvFont font;

cvInitFont(&font, CV_FONT_HERSHEY_SIMPLEX, 1.0, 1.0, 0, 1, CV_AA);

cvPutText(img, "Hello ^_^", cvPoint(40, 40), &font, cvScalar(255, 0, 0, 0));

/* display the image */

cvNamedWindow("image", CV_WINDOW_AUTOSIZE);

cvShowImage("image", img);

cvWaitKey(0);

cvDestroyWindow("image");
cvReleaseImage( &img );
return 0;
}

The output is:

The command used initiating font ,size ,thickness is of format:

cvInitFont(CvFont* font, int fontFace, double hscale, double vscale, double shear=0, int thickness=1, int lineType=8)

The command cvPutText() is used to wite the text on the image at required coordinates.
cvPutText(CvArr* img, const char* text, CvPoint org, const CvFont* font, CvScalar color)

Draws a text string.

Parameters:	img – Input image text – String to print org – Coordinates of the bottom-left corner of the first letter font – Pointer to the font structure color – Text color

Color and shape based tracking of a ball

2011-05-13T11:16:00.000-07:00

In image processing, there are various color models like RGB, HSV, HSI, CMYK. For image proccessing regaurding color detection, RGB and HSV models are commonly used.

The HSV( Hue Saturation Value ) color model solves our purpose of color detection best. In RGB, to define a color, the ranges of red, blue and green must be defined, but in HSV model, for a color we need to define only the value of Hue, which is defined as the color descriptor, or the parameter for wavelength describing a particular color. The saturation is defined as the prominence of hue, and value is a parameter dependent upon the lighting conditions.

There are ranges of Hue values fixed for different colors.These values can be seen on
http://mkweb.bcgsc.ca/color_summarizer/?faq

We would be needing a software colorpic to check the hue saturation and value from images. It can be downloaded from
http://www.iconico.com/colorpic/

To detect a color ball, we need to go through color and then shape detection

STEP1: Capture a frame from the camera

CvCapture* capture = cvCaptureFromCAM( 0 ); //0 is device id
IplImage* frame = cvQueryFrame( capture ); //frame is captured

cvQueryFrame() is used to capture a frame from camera of device id specied in the object capture.

STEP2: Convert the color model of image from RGB to HSV

cvCvtColor(frame, hsv_frame, CV_BGR2HSV);
It coverts the color model of the RGB image frame to HSV and stores it in hsv_frame without affecting frame.the format of this command is:

cvCvtColor(imginit, imgfinal, color space conversion);

where,        imginit=image whose color space is to be changed
                  imgfinal=modified color space image
                  color space conversion =CV_BGR2HSV //for rgb to hsv
                                                      =CV_HSV2BGR//for hsv to rgb

STEP3: Specify the color to be detected by asigning the minimum and maximum values for hue saturation and value

CvScalar hsv_min = cvScalar(35, 80, 170, 0);             //min value of h, s and v
CvScalar hsv_max = cvScalar(55, 130, 255, 0);         //max value of h, s and v

This is in accordance with the color to be detected. It is also highly dependent on lighting conditions. The values of Hue Saturation and Value can be checked by using the software colorpic.

STEP4: Color detection

cvInRangeS(hsv_frame, hsv_min, hsv_max, thresholded);
This command checks for the regions in the image having the values of h, s and v within the permissible range, i.e. between hsv_min and hsv_max, and the binary image thus formed is stored in thresholded.

STEP5: Shape detection i.e. detection of circle

cvSmooth( thresholded, thresholded, CV_GAUSSIAN, 11, 11 );   //smoothing of image

This command is used to smooth a image using gaussian smoothing using a square of size 11x11. The area can range from 0x0 to 11x11, depending on whether the smoothing we require is used to detect small objects or not. The greater the area the better the detection of large objects.

         CvSeq* circles = cvHoughCircles(thresholded, storage, CV_HOUGH_GRADIENT, 2,thresholded->height/6, 100, 50, 10, 400);

This command is used to detect circle shape in the binary image thresholded. The format of the command is:

cvHoughCircles(IntPtr image,IntPtr circleStorage,HOUGH_TYPE method,double dp,double minDist,double param1,double param2,int minRadius,int maxRadius)

STEP6: Draw a circle of desired color around the detected circle

cvCircle(IntPtr img,Point center,int radius,MCvScalar color,int thickness,LINE_TYPE lineType,int shift)

It is used to draw a circle of radius r at centre x,y.

Now, you are ready to detect a color ball. :)

The complete code is:

#include <cvaux.h>
#include <highgui.h>
#include <cxcore.h>
#include <stdio.h>
#include<iostream>
#include <stdlib.h>
#include <string.h>
#include <assert.h>
#include <math.h>
#include <float.h>
#include <limits.h>
#include <time.h>
#include <ctype.h>
using namespace std;
int main(int argc, char* argv[])
{
IplImage* frame = NULL;

     CvCapture* capture = cvCaptureFromCAM( 0 );

     if( !capture )
     {
             fprintf( stderr, "ERROR: capture is NULL \n" );
             getchar();
             return -1;
     }

    cvNamedWindow("video", CV_WINDOW_AUTOSIZE);
    cvNamedWindow("colordetect", CV_WINDOW_AUTOSIZE);

    CvScalar hsv_min = cvScalar(35, 80, 170, 0);
    CvScalar hsv_max = cvScalar(55, 130, 255, 0);

    frame = cvQueryFrame( capture );
    CvSize size = cvGetSize(frame);

     IplImage* hsv_frame    = cvCreateImage(size, IPL_DEPTH_8U, 3);

     IplImage* thresholded   = cvCreateImage(size, IPL_DEPTH_8U, 1);

while(1)
     {

        //capture image from cam
         frame = cvQueryFrame( capture );

         if( !frame )
         {
                 fprintf( stderr, "ERROR: frame is null...\n" );
                 getchar();
                 break;
         }

        //conversion of color model from rgb to hsv
        cvCvtColor(frame, hsv_frame, CV_BGR2HSV);

        //color detection
        cvInRangeS(hsv_frame, hsv_min, hsv_max, thresholded);

         // Memory for hough circles
         CvMemStorage* storage = cvCreateMemStorage(0);

        //Smoothing of image as cvHoughCircles works better on smoothed image
         cvSmooth( thresholded, thresholded, CV_GAUSSIAN, 11, 11 );

        //shape detection-circle shape
         CvSeq* circles = cvHoughCircles(thresholded, storage, CV_HOUGH_GRADIENT, 2,thresholded->height/6, 100, 50, 10, 400);

         cout<<"no. of circles="<<circles->total;

         //drawing circle around the detected circle
         for (int i = 0; i < circles->total; i++)
         {
             float* p = (float*)cvGetSeqElem( circles, i );
             cout<<"x= "<<p[0]<<"y= "<<p[1]<<"r= "<<p[2]<<endl;
             cvCircle( frame, cvPoint(cvRound(p[0]),cvRound(p[1])),cvRound(p[2]), CV_RGB(255,0,0), 3, 8, 0 );
         }

    //display image showing ball detected
cvShowImage("video", frame);

    //color detected image
    cvShowImage("colordetect", thresholded);

    //release memry storage
    cvReleaseMemStorage(&storage);

    //wait for 10 sec and continue or break if "ESC"is pressed
    if( (cvWaitKey(10) & 255) == 27 ) break;
     }
      //release capture
      cvReleaseCapture( &capture );

      cvDestroyWindow( "mywindow" );
      return 0;
    }

Take Video input from webcam and store it as a video file

2011-05-11T08:37:00.000-07:00

In the previous article we discussed how to install OpenCV and how to read an image and display it.
Now, let’s go through the code regaurding how to take input from webcam and save it to a video file.
A video is a collection of images captured so fast that we cannot recognise the difference due to persistence of vision.

Code:

#include <opencv/cvaux.h>
#include <opencv/highgui.h>
#include <opencv/cxcore.h>
#include <stdio.h>
#include<iostream>
using namespace std;
int main(int argc, char* argv[])
{
int coun=0;
int camera_index = 0;
IplImage* frame = NULL;
     CvCapture* capture = cvCaptureFromCAM( 0 );
     if( !capture )
     {
             fprintf( stderr, "ERROR: capture is NULL \n" );
             getchar();
             return -1;
     }

     double cam_w = cvGetCaptureProperty(capture, CV_CAP_PROP_FRAME_WIDTH);
      double cam_h = cvGetCaptureProperty(capture, CV_CAP_PROP_FRAME_HEIGHT);
      printf("* Capture properties: %f x %f - %f fps", cam_w, cam_h, 18);

    cvNamedWindow("video", CV_WINDOW_AUTOSIZE);
    CvVideoWriter* writer = NULL;
    writer = cvCreateVideoWriter("out.avi", CV_FOURCC_DEFAULT, 18, cvSize((int)cam_w,(int)cam_h), 1);
    //Check for creation of writer
    if (writer == NULL)
     {
      printf("!!! ERROR: cvCreateVideoWriter");
      return -1;
     }
while(1)
     {
        frame = cvQueryFrame( capture );
         if( !frame )
         {
                 fprintf( stderr, "ERROR: frame is null...\n" );
                 getchar();
                 break;
         }
cvShowImage("video", frame);
cvWriteFrame(writer, frame);
         if( (cvWaitKey(10) & 255) == 27 ) break;
     }
      cvReleaseVideoWriter(&writer);
      cvReleaseCapture( &capture );
      cvDestroyWindow( "mywindow" );
      return 0;
    }

Explanation:

CvCapture* capture = cvCaptureFromCAM( 0 );

This function takes as its argument the device ID of camera from which video is to be captured and then returns a pointer to a CvCapture structure. This structure contains all of the information about the video. When created in this way, the CvCapture structure is initialized to the beginning of video input capture.

writer = cvCreateVideoWriter("out.avi",CV_FOURCC_DEFAULT,18, cvSize((int)cam_w,(int)cam_h),1);

This function is used to create a writer to write the captured video to the video file "out.avi".

frame = cvQueryFrame( capture );

Once inside of the while(1) loop, we begin taking input from the camera.

if( (cvWaitKey(10) & 255) == 27 ) break;
Here the code waits for 10 ms and continues execution until "Esc" is pressed.

cvReleaseVideoWriter(&writer);
Because there was no more video data or because the user hit the Esc key, after the video has been successfully written to "out.avi", we need to release the writer.

cvReleaseCapture( &capture );
When we have exited the read loop (because there was no more video data or because the user hit the Esc key) we can free the memory associated with the CvCapture structure with the above function.

Now if only the processing of video is required and writting to video file is not required ,yhen we can remove the writer and code will be as follows:

#include <opencv/cvaux.h>
#include <opencv/highgui.h>
#include <opencv/cxcore.h>
#include <stdio.h>
#include<iostream>
using namespace std;
int main()
{
    IplImage* frame = NULL;
     CvCapture* capture = cvCaptureFromCAM( 0 );
     if( !capture )
     {
             fprintf( stderr, "ERROR: capture is NULL \n" );
             getchar();
             return -1;
     }

cvNamedWindow("video", CV_WINDOW_AUTOSIZE);

     while(1)
     {
        frame = cvQueryFrame( capture );
         if( !frame )
         {
                 fprintf( stderr, "ERROR: frame is null...\n" );
                 getchar();
                 break;
         }
cvShowImage("video", frame);

if( (cvWaitKey(10) & 255) == 27 ) break;
     }
cvReleaseCapture( &capture );
      cvDestroyWindow( "mywindow" );
return 0;
}

Introduction to OpenCV for Image Processing

2011-05-11T05:24:00.000-07:00

OpenCV (Open Source Computer Vision) is a computer vision library available from [http://opencvlibrary.sourceforge.net/]. It is an open source library written in C and C++ and runs under Windows, Linux and Mac OS X. OpenCV has lot of inbuilt functions that you can use for your image processing and computer vision needs. It is mainly aimed at real time applications.

In this article,I will be telling u how to install OpenCV on your computer and integrating with Code::Blocks IDE so that you can say ‘hello’ to the world of image processing.

The latest version of OpenCV can be downloaded from http://sourceforge.net/projects/opencvlibrary/files/opencv-doc/Tutorials%2C%20Presentations/

Configuring Code::Blocks and integrating OpenCV

STEP1: From the main menu select: Settings -> Compiler and debugger
Set the Selected Compiler to GNU GCC Compiler

In the window select, Global compiler settings and go to the tab, Linker Settings
Click Add and add all the following directories one by one

C:\OpenCV2.0\lib\libcv200.dll.a
C:\OpenCV2.0\lib\libcvaux200.dll.a
C:\OpenCV2.0\lib\libcxcore200.dll.a
C:\OpenCV2.0\lib\libcxts200.dll.a
C:\OpenCV2.0\lib\libhighgui200.dll.a
C:\OpenCV2.0\lib\libml200.dll.a

In the window select, Global compiler settings and go to the tab, Search directories -> Compiler
Click Add and add all the following directories one by one

C:\OpenCV2.0\include
C:\OpenCV2.0\include\opencv

Now go to the Linker tab (next to the Compiler tab) and add the following directory

C:\OpenCV2.0\lib

STEP2: In the header file cxoperations.hpp ,edit it by adding __MINGW32__ to the end of line number 68 or replace the line by the line given below.
   #if __GNUC__ >= 4 || __MINGW32__

Now, you have completed the installation of OpenCV and are good to go.

Lets start with a basic program like reading an image and displaying it.

//Headers to include opencv library
#include <cv.h>
#include <cxcore.h>
#include <highgui.h>
int main()
{

IplImage* img = cvLoadImage("E:/walls/014.jpg");    //load an image. Note the "/" is used.

cvNamedWindow("Image", CV_WINDOW_AUTOSIZE);  //Create window to display image

cvShowImage("Image", img);       //View image

cvWaitKey(0);                           //Program waits till eternity for a key to be pressed.

cvDestroyWindow("Image");            //Destroy window

cvReleaseImage(&img);                   //Release image

return 0;

}

The image will be displayed till a key is pressed.
Congrats! u have just written ur 1st OpenCV program. :)

In case of any queries, contact me at my mail:- nayakamitarup@gmail.com or reply on the post.