Nimishan - Views and Thoughts

Monday, June 2, 2014

Viola–Jones object detection framework

The Viola–Jones object detection framework is the first object detection framework to provide competitive object detection rates in real-time proposed in 2001 by Paul Viola and Michael Jones. Although it can be trained to detect a variety of object classes, it was motivated primarily by the problem of face detection. This algorithm is implemented in OpenCV as cvHaarDetectObjects().

Feature types and evaluation

The feature employed by the detection framework universally involve the sums of image pixels within rectangular areas. As such, they bear some resemblance to Haar basis functions, which have been used previously in the realm of image-based object detection. However, since the features used by Viola and Jones all rely on more than one rectangular area, they are generally more complex. The figure at right illustrates the four different types of features used in the framework. The value of any given feature is always simply the sum of the pixels within clear rectangles subtracted from the sum of the pixels within shaded rectangles. As is to be expected, rectangular features of this sort are rather primitive when compared to alternatives such as steerable filters. Although they are sensitive to vertical and horizontal features, their feedback is considerably coarser. However, with the use of an image representation called the integral image, rectangular features can be evaluated in constant time, which gives them a considerable speed advantage over their more sophisticated relatives. Because each rectangular area in a feature is always adjacent to at least one other rectangle, it follows that any two-rectangle feature can be computed in six array references, any three-rectangle feature in eight, and any four-rectangle feature in just nine.

Learning algorithm

The speed with which features may be evaluated does not adequately compensate for their number, however. For example, in a standard 24x24 pixel sub-window, there are 162,336 possible features in total, and it would be prohibitively expensive to evaluate them all. Thus, the object detection framework employs a variant of the learning algorithm AdaBoost to both select the best features and to train classifiers that use them.

Download Principles of Digital Image Processing

by: Nimishan No comments

categories: BOOKS, DIGITAL IMAGE PROCESSING

Download -Digital-Image-Processing---Fundamental-Techniques.pdf

C++ OpenGL Lesson 1: A Simple OpenGL Project

by: Nimishan No comments

categories: Opengl

Advantages & Disadvantages of keys in Criptography

by: Nimishan No comments

categories: Criptography

Symmetric Key Encryption
Symmetric key encryption is also known as shared-key, single-key, secret-key, and private-key or one-key encryption. In this type of message encryption, both sender and receiver share the same key which is used to both encrypt and decrypt messages. Sender and receiver only have to specify the shared key in the beginning and then they can begin to encrypt and decrypt messages between them using that key. Examples include AES (Advanced Encryption Standard) and TripleDES (Data Encryption Standard).
Advantages
- Simple: This type of encryption is easy to carry out. All users have to do is specify and share the secret key and then begin to encrypt and decrypt messages.
- Encrypt and decrypt your own files: If you use encryption for messages or files which you alone intend to access, there is no need to create different keys. Single-key encryption is best for this.
- Fast: Symmetric key encryption is much faster than asymmetric key encryption.
- Uses less computer resources: Single-key encryption does not require a lot of computer resources when compared to public key encryption.
- Prevents widespread message security compromise: A different secret key is used for communication with every different party. If a key is compromised, only the messages between a particular pair of sender and receiver are affected. Communications with other people are still secure.
Disadvantages
- Need for secure channel for secret key exchange: Sharing the secret key in the beginning is a problem in symmetric key encryption. It has to be exchanged in a way that ensures it remains secret.
- Too many keys: A new shared key has to be generated for communication with every different party. This creates a problem with managing and ensuring the security of all these keys.
- Origin and authenticity of message cannot be guaranteed: Since both sender and receiver use the same key, messages cannot be verified to have come from a particular user. This may be a problem if there is a dispute.
Asymmetric/Public Key Encryption
Also known as public key encryption, this method of encrypting messages makes use of two keys: a public key and a private key.The public key is made publicly available and is used to encrypt messages by anyone who wishes to send a message to the person that the key belongs to. The private key is kept secret and is used to decrypt received messages. An example of asymmetric key encryption system is RSA.
Advantages
- Convenience: It solves the problem of distributing the key for encryption.Everyone publishes their public keys and private keys are kept secret.
- Provides for message authentication: Public key encryption allows the use of digital signatures which enables the recipient of a message to verify that the message is truly from a particular sender.
- Detection of tampering: The use of digital signatures in public key encryption allows the receiver to detect if the message was altered in transit. A digitally signed message cannot be modified without invalidating the signature.
- Provide for non-repudiation: Digitally signing a message is akin to physically signing a document. It is an acknowledgement of the message and thus, the sender cannot deny it.

Disadvantages
- Public keys should/must be authenticated: No one can be absolutely sure that a public key belongs to the person it specifies and so everyone must verify that their public keys belong to them.
- Slow: Public key encryption is slow compared to symmetric encryption. Not feasible for use in decrypting bulk messages.
- Uses up more computer resources: It requires a lot more computer supplies compared to single-key encryption.
- Widespread security compromise is possible: If an attacker determines a person's private key, his or her entire messages can be read.
- Loss of private key may be irreparable: The loss of a private key means that all received messages cannot be decrypted.

What is a Digital Signature?

by: Nimishan No comments

categories: Criptography

What is a Digital Signature? An introduction to Digital Signatures, by David Youd

Bob

(Bob's public key)

(Bob's private key)

Bob has been given two keys. One of Bob's keys is called a Public Key, the other is called a Private Key.

Bob's Co-workers:
			Anyone can get Bob's Public Key, but Bob keeps his Private Key to himself
Pat	Doug	Susan

Bob's Public key is available to anyone who needs it, but he keeps his Private Key to himself. Keys are used to encrypt information. Encrypting information means "scrambling it up", so that only a person with the appropriate key can make it readable again. Either one of Bob's two keys can encrypt data, and the other key can decrypt that data.
Susan (shown below) can encrypt a message using Bob's Public Key. Bob uses his Private Key to decrypt the message. Any of Bob's coworkers might have access to the message Susan encrypted, but without Bob's Private Key, the data is worthless.

"Hey Bob, how about lunch at Taco Bell. I hear they have free refills!"

HNFmsEm6Un BejhhyCGKOK JUxhiygSBCEiC 0QYIh/Hn3xgiK BcyLK1UcYiY lxx2lCFHDC/A

"Hey Bob, how about lunch at Taco Bell. I hear they have free refills!"

With his private key and the right software, Bob can put digital signatures on documents and other data. A digital signature is a "stamp" Bob places on the data which is unique to Bob, and is very difficult to forge. In addition, the signature assures that any changes made to the data that has been signed can not go undetected.

To sign a document, Bob's software will crunch down the data into just a few lines by a process called "hashing". These few lines are called a message digest. (It is not possible to change a message digest back into the original data from which it was created.)

Bob's software then encrypts the message digest with his private key. The result is the digital signature.

Finally, Bob's software appends the digital signature to document. All of the data that was hashed has been signed.

Bob now passes the document on to Pat.

First, Pat's software decrypts the signature (using Bob's public key) changing it back into a message digest. If this worked, then it proves that Bob signed the document, because only Bob has his private key. Pat's software then hashes the document data into a message digest. If the message digest is the same as the message digest created when the signature was decrypted, then Pat knows that the signed data has not been changed.

Plot complication...

Doug (our disgruntled employee) wishes to deceive Pat. Doug makes sure that Pat receives a signed message and a public key that appears to belong to Bob. Unbeknownst to Pat, Doug deceitfully sent a key pair he created using Bob's name. Short of receiving Bob's public key from him in person, how can Pat be sure that Bob's public key is authentic?

It just so happens that Susan works at the company's certificate authority center. Susan can create a digital certificate for Bob simply by signing Bob's public key as well as some information about Bob.

Bob Info:
    Name
    Department
    Cubical Number Certificate Info:
    Expiration Date
    Serial Number Bob's Public Key:

Now Bob's co-workers can check Bob's trusted certificate to make sure that his public key truly belongs to him. In fact, no one at Bob's company accepts a signature for which there does not exist a certificate generated by Susan. This gives Susan the power to revoke signatures if private keys are compromised, or no longer needed. There are even more widely accepted certificate authorities that certify Susan.
Let's say that Bob sends a signed document to Pat. To verify the signature on the document, Pat's software first uses Susan's (the certificate authority's) public key to check the signature on Bob's certificate. Successful de-encryption of the certificate proves that Susan created it. After the certificate is de-encrypted, Pat's software can check if Bob is in good standing with the certificate authority and that all of the certificate information concerning Bob's identity has not been altered.
Pat's software then takes Bob's public key from the certificate and uses it to check Bob's signature. If Bob's public key de-encrypts the signature successfully, then Pat is assured that the signature was created using Bob's private key, for Susan has certified the matching public key. And of course, if the signature is valid, then we know that Doug didn't try to change the signed content.