An Effective Optimization Technique of Extracted Watermark Image Using Particle Swarm Optimisation

: Nowadays the presence of noises in the extracted image is a major constrain in the Image Processing environment. An ultimate goal of this paper is to optimize the watermark image using particle swarm optimization. Earlier, the research focused on the video watermarking process where the watermark image is embedded into the video as invisible and can be sent to the receiver side for an extraction. While at the extraction stage, the image what embedded was not been retrieved at the receiver side due to the presence of various noises and occurance of errors in the medium. Hence while comparing the extracted image with the original image, the accuracy was poor and non reconstructable. So this paper serves solution for the above said problem using Particle Swarm Optimization Technique. As the end, the extracted image can be reconstructed to the maximum extent as equal to the input image. This scheme can be widely used in the Medical Imaging application, Defence, etc.


A. INTRODUCTION
Particle Swarm Optimization (PSO) is a sturdy technique based on the drive and intelligence of swarms. This Optimization technique applies the concept of social interaction to problem solving techniques. This Optimization Technique was developed by James Kennedy and Russell Eberhart in the year 1995.
PSO is considered to be an efficient optimization algorithm by searhcing an entire highdimensional problem space. It uses number of agents considered as particles; that constitute a swarm moving around in the search space looking for the best solution. Each particle is considered as a point in an N dimensional space which adjusts as a flying experince of other particles. In particle swarm optimization, if a member finds a desirable path, the rest will be followed by the swarms. Each member learns not only from its own experience but from others especially from the best performer. As of the Particle Swarm Optimization, there are two variants which are Basic Variant Particle Swarm Optimization and Modification variant Particle Swarm Optimization. The basics of Particle Swarm Optimization include: The standard version of the PSO algorithm is essentially described by the following two simple velocity and position update equations, shown in 1 and 2 respectively. Vid(t+1) = W Vid(t) + C1R1(Pid(t) -Xid(t)) + C2R2(Gbid(t) -Xid(t)) ---(1) and Xid(t+1) = Xid(t) + Vid(t+1) ---(2) Where, Vid → Rate of position change (velocity).
Xid → Position of the i th particle in the d th dimension. Pid → Historically best position of the i th particle in the d th dimension. R1 and R2 → ndimensional vectors. c1 and c2 → Cognitive and Social parameters.

B. WORK FLOW OF THE PROJECT
Step 1: Initialize the particles.
Step 2: Calculate the fitness value for each particles.
Step 3: If fitness value is better than the particles best fitness value (Pbest) in the history set current value as the new Pbest.
Step 4: Repeat step 2 and 3 for all the particles.
Step 5: Choose the particle with best fitness value compared to all the particles and label as the global best (gbest).
Step 6: Calculate and update each particles velocity and position for all the particles.
Step 7: Calculate the fitness value for each particles.
Step 8: If a minimum error criterion ((i.e.) desired fitness value of 90% of all the particles) is not achieved, goto step 3.
Step 9: Display the particles.
The following diagram represents the research flow graph where the particle swarm optimization undergoes various level of updation inorder to get the optimized swarm output. Initially, the process initializes the supersparm where the image consists of multiswarm in which the Supersprams are identified. Those superswarms are considered to be the super sized swarm. Then the process is continued to evaluate the fitness of each particles of the swarms. This inturn measures the dimension of the particle or swarm www.psychologyandeducation.net parameters. The next process is to initialize all the subswarms following to the superswarms and then the object functions are evaluated inorder to get the fitness value of each particle or swarm. Once the process is completed, the updation of particles were made and continued to the further movement. Similarly, the same process is repeated for the swarm parameters too as a separate condition. Hence this process serves as the main work flow in enhancing the quality of the swarms using Particle Swarm Optimization.

General Objective
The general objective of the research is to enhance the quality of an image using Particle Swarm Optimization.

Specific Objective
The specific objective of the research includes: 1. To improve the interpretability of an image. 2. To improve image perception. 3. To reduce the noise occurrence.

D. BLOCK DIAGRAM
In this method, the local gain parameters were used. Along with the above said parameters, the problems with other image enhancement techniques were overcome.

E. SIMULATION RESULT
The proposed algorithm has been tested with various numbers of Logos, medical images and images downloaded from the internet databases. Based on the various images, the following result has been obtained. In this section, the various performance of the image enhancement method is discussed over. The parameters taken into the considerations are Peak Signal to Noise Ratio (PSNR) and Bit Correction Rate (BCR). And the Simulation Result for the proposed algorithm stands the following table which is prescribed in the Table 1. As like, it is found to be an improved PSNR for all the cases when the images are submerged into various attacks and noises. Also, it is found that the Bit Correction Rate for all the images with PSO seems to be quiet impressive and improved.