Papers by Rahul Kumar
Macromolecular Symposia, Jan 23, 2015
The finger print recognition, face recognition, hand geometry, iris recognition, voice scan, sign... more The finger print recognition, face recognition, hand geometry, iris recognition, voice scan, signature, retina scan and several other biometric patterns are being used for recognition of an individual. Human footprint is one of the relatively new physiological biometrics due to its stable and unique characteristics. The texture and foot shape information of footprint offers one of the powerful means in personal recognition. This work proposes a footprint based biometric identification of an individual by extracting texture and shape based features using Principal Component Analysis (PCA) and Independent Component Analysis (ICA) linear projection techniques. PCA is a commonly used technique for data classification and dimensionality reduction and ICA is one of the most widely used blind source separation technique for revealing hidden factors that underlie sets of random variables, measurements, or signals. In this study PCA and ICA have been compared for footprint recognition using distance classification techniques such as Euclidean distance, city block, cosine and correlation. Experimental results show that ICA performs better than PCA for footprint recognition.
2nd International Conference on Research in Science, Engineering and Technology (ICRSET’2014), Mar 22, 2014
This paper proposes an algorithm for biometric palm recognition of an individual using geometrica... more This paper proposes an algorithm for biometric palm recognition of an individual using geometrical and texture properties. Biometrics is physiological characteristics of human beings, unique for every individual that are usually time invariant & easy to acquire. Palmprint is one of the physiological biometrics due to its stable & unique characteristics. The physical dimensions of a human palm contains information that is capable of authenticating the identity of an individual. In this study algorithm is proposed to extract a finger length as geometrical feature and principal lines as texture feature. Simulation results shows false accept rate (FAR) is 25%, false reject rate (FRR) as 18.75%, genuine acceptance rate (GAR) 81.25%, half total error rate (HTER) 21.87% and accuracy 78.12% in case of geometrical features whereas accuracy is 92.3% in case of texture feature.
Human palm print is a wide spreading biometric trait that has been used to detect an individual’s... more Human palm print is a wide spreading biometric trait that has been used to detect an individual’s identity Palm print images of 94 individuals have been acquired using Scanning techniques. Images acquired by this technique are used for the identification of an individual by extracting principle lines as a feature of palm print images using Matlab. In this study an algorithm is proposed to compute and filter the first three principle lines of palm print images using Matlab that have been acquired using scanning technique. These characteristic features are used to estimate the identity of an individual. Experiments show that the principle lines from the palm prints can be a good source for any biometric security system.
The International Journal of Advancements in Electronics and Electrical Engineering
This paper proposes an algorithm for authentication of an individual using palm prints... more This paper proposes an algorithm for authentication of an individual using palm prints.
Biometrics is physiological characteristics of human beings, unique for every individual that
are usually time invariant & easy to acquire. Palmprint is one of the physiological biometrics
due to its stable & unique characteristics. The physical dimensions of a human palm contains
information that is capable of authenticating the identity of an individual. In this study
algorithm is proposed to compute 512-bit security key using figure length of fingers of an
individual. The performance of a key is assessed on1280 keys stored in a database. Simulation
results shows false accept rate (FAR) is 25%, false reject rate (FRR) as 18.75%, genuine
acceptance rate (GAR) 81.25%, half total error rate(HTER) 21.87% and accuracy 78.12%.
In this report we are going to discuss about the learning aspects of an antivirus for making appr... more In this report we are going to discuss about the learning aspects of an antivirus for making appropriate decisions using decision tree. Combining such techniques we can improve the performance of the Artificial Intelligent Algorithms for the virus detection and prevention.
Conference Presentations by Rahul Kumar
The International Journal of Advancements in Electronics and Electrical Engineering
This paper proposes an algorithm for authentication of an individual using palm prints... more This paper proposes an algorithm for authentication of an individual using palm prints.
Biometrics is physiological characteristics of human beings, unique for every individual that
are usually time invariant & easy to acquire. Palmprint is one of the physiological biometrics
due to its stable & unique characteristics. The physical dimensions of a human palm contains
information that is capable of authenticating the identity of an individual. In this study
algorithm is proposed to compute 512-bit security key using figure length of fingers of an
individual. The performance of a key is assessed on1280 keys stored in a database. Simulation
results shows false accept rate (FAR) is 25%, false reject rate (FRR) as 18.75%, genuine
acceptance rate (GAR) 81.25%, half total error rate(HTER) 21.87% and accuracy 78.12%.
Uploads
Papers by Rahul Kumar
Biometrics is physiological characteristics of human beings, unique for every individual that
are usually time invariant & easy to acquire. Palmprint is one of the physiological biometrics
due to its stable & unique characteristics. The physical dimensions of a human palm contains
information that is capable of authenticating the identity of an individual. In this study
algorithm is proposed to compute 512-bit security key using figure length of fingers of an
individual. The performance of a key is assessed on1280 keys stored in a database. Simulation
results shows false accept rate (FAR) is 25%, false reject rate (FRR) as 18.75%, genuine
acceptance rate (GAR) 81.25%, half total error rate(HTER) 21.87% and accuracy 78.12%.
Conference Presentations by Rahul Kumar
Biometrics is physiological characteristics of human beings, unique for every individual that
are usually time invariant & easy to acquire. Palmprint is one of the physiological biometrics
due to its stable & unique characteristics. The physical dimensions of a human palm contains
information that is capable of authenticating the identity of an individual. In this study
algorithm is proposed to compute 512-bit security key using figure length of fingers of an
individual. The performance of a key is assessed on1280 keys stored in a database. Simulation
results shows false accept rate (FAR) is 25%, false reject rate (FRR) as 18.75%, genuine
acceptance rate (GAR) 81.25%, half total error rate(HTER) 21.87% and accuracy 78.12%.