International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 10 Issue: 05 | May 2023
p-ISSN: 2395-0072
www.irjet.net
Design and Implementation of JPEG CODEC using NoC MADIHA KAUSAR1, MOHAMMED ARSHAD MURTAZA2, KAUSER ANJUM3 1Assistant Professor, Dept. of E&CE, K.C.T. Engineeering College, Kalaburagi, Karnataka, India
2M.Tech Student, VLSI & Embedded Systems, Sharnbasva University, Kalaburagi, Karnataka, India
Ph.D, Dept. of E&CE, K.C.T. Engineeering College, Kalaburagi, Karnataka, India ---------------------------------------------------------------------***--------------------------------------------------------------------challenging or impossible to carry out in the usual Abstract - An image compression and decompression 3
analogue format, the digital mode has largely supplanted analogue technology in various applications. So, both when they are created and when they are presented, video signals are intrinsically analogue. There is an opportunity to transform these signals into digital format in between these end points.
system using NoC structure is developed and implemented. Due to the increasing demands of image transmission in computer and mobile systems, research on image compression has increased significantly. The JPEG (Joint Photographic Experts Group) standard is a commonly used method for lossless compression of digital images, especially those produced by digital photography. Digital images require a lot of storage space. The aim of this work is to design a VLSI architecture for basic sequential JPEG encoding and decoding processes. A codec that specifies how an image is compressed into a stream of bytes and then decompressed back into an image is defined by the JPEG standard. To achieve high execution speed, the architecture makes best use of pipeline and parallelism principles. Providing effective on-chip communication networks, network-on-chip (NoC) is a new paradigm in complicated system-on-chip (SoC) designs. It enables scalable communication and decoupling of communication and computing. Data is sent over networks as packets. Data routing is primarily done through routers. The router architecture must therefore be effective, with lower latency and higher throughput; the effectiveness of the router architecture is assessed in a two-dimensional network topology. NoC is used to speed up the image transfer process and thus provide a shorter processing time for transferring data over the network. The design is implemented using Verilog HDL and simulated in Xilinx ISE Design Suite 14.7.
For effective storage and transmission, digital images must be compressed because they have a significant amount of data. The need for digital picture storage, editing, and transport has exploded with the advent of digital cameras. These image files have a tendency to be enormous and take up a lot of RAM. A typical 640 x 480 color image has about a million elements, but a 256 x 256 grayscale image has 65,536 to store. Standard broadcast television requires 100 to 200 Mbps for video applications, whereas low resolution applications like teleconferencing, remote surveillance, etc. just need a few megabits per second, when the frame rate and word size are decreased, the data rate can be brought down to a few tens of kilobits. Using picture compression methods at the source could result in even more compression. At the destination, these compressed images are decompressed for viewing and analysis. Many picture compression methods have been developed in the recent years. These methods seek to reduce the size of the photos by sending just the non-redundant data as it is received. Other methods are inherently lossy. In this case, the reconstructed images are actually close copies of the uncompressed originals. The original and reconstructed images' corresponding pixel values are different. As a result, compression may be accomplished without significantly affecting the visual image quality. A group of picture compression mechanisms are defined by the JPEG standard. The baseline sequential encoding scheme is the mechanism that is most frequently utilised. This approach for lossy compression [1] is used.
Key Words: Image processing, Lossy Image compression, Image decompression, JPEG, CODEC, NoC.
1. INTRODUCTION The reason why digital representation of the signal is more reliable than its analogue counterpart for processing, manipulation, storage, recovery, and transmission over long distances, even across the globe through communication networks, is why we are talking about digital networks, digital representation of images, movies, video, TV, voice, and digital libraries today. Digital technology has only recently been used to the transmission, recording, and processing of images, but it has quickly revolutionised many areas of image processing. Because it can carry out operations that are
© 2023, IRJET
|
Impact Factor value: 8.226
Image compression and decompression techniques can be used to solve the digital picture transmission issue. In order to speed up the forwarding of images and provide a shorter processing time to transfer data through the network with a low latency, it can only transmit nonredundant data, as it does with NoC. Providing effective on-chip communication networks, a network-on-chip
|
ISO 9001:2008 Certified Journal
|
Page 1762