# 3- idiots part-3

Views:

Category: Entertainment

## Presentation Description

No description available.

## Presentation Transcript

### Slide 1:

SURVEY ON RESEARCH FIELD “IMAGE PROCESSING” RESEARCH TOPIC “IMPLEMENTING FACE RECOGNITION USING A PARALLEL.. IMAGE PROCESSING ENVIRONMENT BASED ON ALGORITHMIC SKELETONS” MASTER OF COMPUTER APPLICATION (4th Semester) Researched by: ANOOP GANGWAR ALOK GUPTA NITISH SHARMA SRMSCET , BAREILLY THIRD PHASE

### Slide 2:

What is Digital image processing? An image may be defined as a two dimensional function F(x , y) ,where x, y are Spatial co-ordinates and the amplitude of F at any pair of the co-ordinates (x , y) is called the intensity . When x , y and amplitude values of F are all finite discrete quantities , then we call the image as the digital image y x

### Slide 3:

What is Face Detection?

### Slide 4:

What is Face Recognition? Source Result

### Slide 5:

How Facial Recognition Systems Work Every face has numerous, distinguishable landmarks, the different peaks and valleys that make up facial features. FaceIt defines these landmarks as nodal points. Each human face has approximately 80 nodal points. Some of these measured by the software are: Distance between the eyes Width of the nose Depth of the eye sockets The shape of the cheekbones The length of the jaw line

### Slide 6:

Difference between Face Detection and Recognition Detection – two-class classification Face vs. Non-face Recognition – multi-class classification One person vs. all the others

### Slide 7:

Applications of Face Recognition Multimedia Management Security Smart Cards Surveillance Others

SECOND PHASE

### Slide 9:

Low-level image operations Low-level image processing operations use the values of image pixels to modify individual pixels in an image. They can be divided into point-to point, neighborhood-to-point and global-to-point operations . Intermediate-level image operations Intermediate-level image processing operations work on images and output other data structures, such as detected objects (e.g., faces) or statistics, thereby reducing the amount of information. Operations such as Hough transform (to find a line in an image), center-of gravity calculation, labeling an object, are examples of intermediate-level image operations. High-level image operations High-level image processing operations work on vector data or objects in the image and return other vector data or objects. They usually have irregular access patterns and thus are difficult to run data parallel. They can be divided into object-to-object or object-to-point operations. Position estimation and object recognition theory are examples of this category.

### Slide 10:

U V 127 127 -127 -127 Skin-Tone Region { Skin region in UV spectrum }

### Slide 11:

U V 127 127 -127 -127 { Skin region in UV spectrum }

### Neural Network Face Detector :

Neural Network Face Detector Input: Color image. Output: P = probability that image contains a face. (Only 1 output node.) Set 1 for face, 0 for no face. P=1 P=0 3 Possible Outputs P > 0.5 FACE P < 0.5 NOT FACE P = 0.5 DON’T KNOW

### What is Face Recognition? :

What is Face Recognition? A set of two task: Face Identification: Given a face image that belongs to a person in a database, tell whose image it is. Face Verification: Given a face image that might not belong to the database, verify whether it is from the person it is claimed to be in the database. 4/6/2010 13

### OUR PROPOSED SOLUTION :

OUR PROPOSED SOLUTION

### CHALLENGES IN IMAGE PROCESSING(FACE DETECTION &RECOGNITION) :

CHALLENGES IN IMAGE PROCESSING(FACE DETECTION &RECOGNITION) Maintainig the quality of an compressed Image. Sure-Shot Question about the Intelligence of machine. Exact capture at rapid motion of image. Reconstrution of noise image. Find faces in an image of same colour

### OUR PROPOSED SOLUTION :

OUR PROPOSED SOLUTION

### Step-1 We WILL DO THIS BY USING A KERNAL,BASED ON MASTER SLAVE TECNIQUE. :

Step-1 We WILL DO THIS BY USING A KERNAL,BASED ON MASTER SLAVE TECNIQUE.

### Design a KERNAL who have certain approaches.. :

Design a KERNAL who have certain approaches.. NURAL NETWORK FOR 3D CONSTRUCTED IMAGE. SKIN COLOR BASED ALGORITHMS. GENETIC ALORITHMS.

### IMAGE DATABASE CONSIST OF….. :

IMAGE DATABASE CONSIST OF….. BUNCH OF IMAGES SUCH AS 10 DIFFERENT IMAGES AS PER FACE….INSTEAD OF SINGLE FACE.

### KERNAL DETECTION APPROACH :

KERNAL DETECTION APPROACH

### 1- AFTER DETECTION A MASTER KERNAL WILL GENERATE THE SAME NUMBER OF SLAVE KERNAL AS THE NUMBER OF FACES DETECTS. :

1- AFTER DETECTION A MASTER KERNAL WILL GENERATE THE SAME NUMBER OF SLAVE KERNAL AS THE NUMBER OF FACES DETECTS. 2- MASTER KERNAL ASSIGN THE WORK TO MONITOR ITS OWN ASSIGNED FACE AND SEND THE REPORT OF RECOGNITION TO MASTER.

### RECOGNITION ALGORITHM.. :

RECOGNITION ALGORITHM.. MASTER KERNALWILL DECIDED THE APPROACH FOR EACH SLAVE KERNALWHICH WILL BEST SUITED AT THAT MOMENT. MASTER KERNAL MONITOR ENTIRE THE PROCESS. IF SLAVE WILL FOUND A VARIYATION IN RECOGNITION GRAPH SEND REPORT TO MASTER AND MASTER TAKE DECESION EITHER CONTENIOUING THE APPROCH OR APPLIED ANOTHER ONE.

### CONT…. :

CONT…. WHEN RECOGNITION PROCESS WILL COMPLETE SLAVE WILL SEND FACE TO MASTER AND MASTER SENDS AGAIN TO THE SLAVE FOR THE PROCESS OF COMPARISION. SLAVE ENTERS IN DATABASE AND MATCH UP THE TESTED IMAGE WITH EACH 13 IMAGES IF MATCH WILL SUCESSFUL THEN IT SENDS RESULT TO MASTER AND MASTER SHOWS OUTPUT TO THE USER.

### EXAMPLE OF “DREAME COME’s TRUE” :

EXAMPLE OF “DREAME COME’s TRUE” TERMINATOR SALVATION

### Slide 29:

TERMINATOR TERMINATOR MIS

THANK YOU

? Any