I see see see Day 2
Keynote Lecture 3:- Distributed Sensor Network - Sitaram Iyengar
The following points were discussed:-
1.Information analysis requires analytic method and variety of sensor types.
2.Sensing the change in modalties - Moore's Law
3.RFID was introduced, RFID industry is over $26million
4.There are two types of sensor networks - Data Gathering and Even Detection.
5.Every node in the network is capable of substantial processing
6.DSN(Distributed Sensor Network)- Uses less energy and small number of nodes are subjected to failure.
7.ADHOC Net:-All mobile nodes can be self organised in an arbitary network.
8.Deployment of Sensors is a continous process.
9.Biochemic sensors for enzymes.
10.Battery life of sensors-real time aggregate of data is a challenge.
11.Syntactic and Semantic processing.
12.A faulty sensor would lead to a fault in a reading which may appear to be normal
13.Use of probability and Bayesian Formula.
Technical Session 4
4.1 Augmented Genetic Algorithm for Auto Corrective Prediction Model for Pest Attacks:- Kumar Saurabh Bisht, Sanjay Chaudhary(DAIICT,Gandhinagar,Gujrat India)
-Prediction model
-Neural Network
-Layer structure of neural network
-Back Propagation Algorithm
-Genetic Algorithm Prediction model
-Database encoded in form of chromosomes.
-His approach has less no. of peaks=>Less errors=>More accuracy -Heuristic Approach
4.2 Biometric Authentication based on Detection and Recognition of Multiple Faces in Image Jolly D. Shah, Prof. S.H.Patil
-Multiple face detection
-Different color models(RGB,CMYK,HSB,YCbCr) were introduced
-The presented technique used YCbCr model.
-Facial Feature:-Distance between eyes calculated (God only knows how !!)
-Then this was converted to grayscale.
-Line of separability model was used.
-Face recognition of criminals
Limitations---Cannot identify identical twins ---False detection ---Update Database regularly --- Aging effect The best combination is - Iris + Face as Biometric Authentication
LUNCH.......YIPPPPPEEEEEEEEE....FREEEEEEDOOOMMMMMM
Keynote 4 - Packet Forwarding - Prof. Sartaj Sahani(University of Florida)
--Simple concepts of networking were discussed. --Data structures like trees and tries were compared. --Ternary CAMS --Difference between dynamic and static routing tables.
Keynote 5 - Natural Language Processing - Rajiv Sangal --DUnnooo kya hua isme... kuch palle hi nahin padha !! Meri language hi kharab ho gayi.. bas !!
TEA BREAK ---- Free Pakoras and Tea --- Got nothing to eat .. all thanx to the 'swimmer' and ppl like him...
Technical Session 5
5.1Collaborative Load Balancing Scheme for Improving Search Performance in Unstructured P2P Networks Sabu M. Thampi, Chandra Sekaran K
5.2Efficient VLSI Layout of Grid Pyramid Networks Saeedeh Bakhshi , Marzieh Bakhshi , Hamid Sarbazi-Azad
Keynote 6 -- Kuch Biotech ka tha... DNA RNA Chromosomes.. ecetra ecetra ecetra
GHAR BHAAGOOOOOOO
The following points were discussed:-
1.Information analysis requires analytic method and variety of sensor types.
2.Sensing the change in modalties - Moore's Law
3.RFID was introduced, RFID industry is over $26million
4.There are two types of sensor networks - Data Gathering and Even Detection.
5.Every node in the network is capable of substantial processing
6.DSN(Distributed Sensor Network)- Uses less energy and small number of nodes are subjected to failure.
7.ADHOC Net:-All mobile nodes can be self organised in an arbitary network.
8.Deployment of Sensors is a continous process.
9.Biochemic sensors for enzymes.
10.Battery life of sensors-real time aggregate of data is a challenge.
11.Syntactic and Semantic processing.
12.A faulty sensor would lead to a fault in a reading which may appear to be normal
13.Use of probability and Bayesian Formula.
Technical Session 4
4.1 Augmented Genetic Algorithm for Auto Corrective Prediction Model for Pest Attacks:- Kumar Saurabh Bisht, Sanjay Chaudhary(DAIICT,Gandhinagar,Gujrat India)
-Prediction model
-Neural Network
-Layer structure of neural network
-Back Propagation Algorithm
-Genetic Algorithm Prediction model
-Database encoded in form of chromosomes.
-His approach has less no. of peaks=>Less errors=>More accuracy -Heuristic Approach
4.2 Biometric Authentication based on Detection and Recognition of Multiple Faces in Image Jolly D. Shah, Prof. S.H.Patil
-Multiple face detection
-Different color models(RGB,CMYK,HSB,YCbCr) were introduced
-The presented technique used YCbCr model.
-Facial Feature:-Distance between eyes calculated (God only knows how !!)
-Then this was converted to grayscale.
-Line of separability model was used.
-Face recognition of criminals
Limitations---Cannot identify identical twins ---False detection ---Update Database regularly --- Aging effect The best combination is - Iris + Face as Biometric Authentication
LUNCH.......YIPPPPPEEEEEEEEE....FREEEEEEDOOOMMMMMM
Keynote 4 - Packet Forwarding - Prof. Sartaj Sahani(University of Florida)
--Simple concepts of networking were discussed. --Data structures like trees and tries were compared. --Ternary CAMS --Difference between dynamic and static routing tables.
Keynote 5 - Natural Language Processing - Rajiv Sangal --DUnnooo kya hua isme... kuch palle hi nahin padha !! Meri language hi kharab ho gayi.. bas !!
TEA BREAK ---- Free Pakoras and Tea --- Got nothing to eat .. all thanx to the 'swimmer' and ppl like him...
Technical Session 5
5.1Collaborative Load Balancing Scheme for Improving Search Performance in Unstructured P2P Networks Sabu M. Thampi, Chandra Sekaran K
5.2Efficient VLSI Layout of Grid Pyramid Networks Saeedeh Bakhshi , Marzieh Bakhshi , Hamid Sarbazi-Azad
Keynote 6 -- Kuch Biotech ka tha... DNA RNA Chromosomes.. ecetra ecetra ecetra
GHAR BHAAGOOOOOOO
Swimmer :P
ReplyDeleteha ha,VJ bhaiya bhi pehchaan gaye,hum bhi"swimmer" ha ha1 kilo pakode toh maare hi honge usne
ReplyDeletenice to see a summary of IC3 :) ... I wish more people at JIIT would have taken interest.. but its a start anyhow.. cheers,Sid
ReplyDelete