Compression minimization of information allows further images to

Compression is a technique that is used to decline
the intensity of a graphical file.JPEG 
is trivial standard that is used for computer images .DCT is a mechanism
that is used to split an image into various sectors of frequencies. Image
Enhancement is a mechanism that is used to alter the image for increasing
contrast, for lessening noise and to smoothen the image so that it produces a
finer image for display. A technique that is used to upgrade the brightness of
an image is called Histogram equalization. In this paper, the JPEG-DCT image
compression scheme with image enhancement is presented to compresses and
decompresses the color images. Image compression and Image Enhancement are two
significant steps that execute the essential methods in image processing.  In the proposed scheme the   image is converted andenhanced using a Histogram equalization
technique and compressed using a JPEG compression algorithm. The proposed
scheme aims   to accomplish greater
compression ratio and finer quality.

Image Compression; JPEG, Histogram Equalization, Discrete Cosine transform.


Image compression is a technique that is used to lessen the size of the
file without diminishing the image quality. This minimization of information
allows further images to be saved in the memory space or disk. Image
compression technique is based on two different categories that are lossy
compression and lossless compression.


Lossy Compression

In this technique the data is lost. After the data has
been compressed the initial data cannot be retrieved or restored. This
technique is used for image and sound compression. MP3 file formats and MPEG
video formats are examples of lossy compression.


In this technique the data is not lost. If data is
compressed, it can be retrieved. Text
compression uses lossless compression. Fax Machine and Radiology are instances
of Lossless Compression.


JPEG is a typical image file format that is standardized by Joint Photographics
Experts Group (usually refers to JPEG) are used in many digital cameras. It is
adopted to compress the decompress the Image. The common file formats of JPEG
are.jpg or .jpeg.In this Paper Compression is carried out on the enhanced image
by the proposed method to achieve larger compression ratio and finer quality.


process that improves the visual perception so that it produces a more relevant
image than that of the original image is called Image Enhancement. The two
methods of Image Enhancement are spatial domain or frequency domain. Spatial
domain techniques are executed on the image pane itself they are based on
explicit use of pixels in an image. Frequency domain techniques enhance the
image by transforming by a continuous position invariant operator. A major
technique of spatial domain is Histogram Equalization (HE) that is used to upgrade
the contrast of the image. HE evenly distributes the values of the pixel within
a range so the output image’s histogram matches with the input image’s
histogram. 3.


standard performs better compared to other previous standards like JPEG-LS, PNG
and MPEG. Where JPEG-LS do not support any purpose like error resilience and scalability,
PNG does not support large image files and lossless coding are not supported in
MPEG. Thus JPEG supports both lossless compression and lossy compression.



Jaffar Iqbal Barbhuiya, Tahera Akhtar Laskar and K. Hemachandran 1 state the
comparison between discrete wavelet transform and discrete cosine transform.
First the image is divided into 8*8 block then the transformation is applied to
the block. Then the DCT Coefficients are quantized using a quantization
technique that reduces the amount of information. The quantized coefficients
are encoded using any of the entropy coding techniques such as Run length
coding, Huffman coding or arithmetic coding. Then again the image is compressed
using discrete wavelet transform. This decomposes the images into three levels
of segments. Then quantization is applied to reduce the amount of information.
Thus quantized coefficients are encoded using several entropy coding
techniques. Thus discrete wavelet transform achieves higher compression ratio
without degrading the quality of the image.

Raid, W.M.Khedr, M. A. El-dosuky and Wesam Ahmed 2 state the JPEG Compression
using DCT. First the image is converted from RGB color space to YUV color
space. Then discrete cosine transform is applied on the image. The results of
DCT are quantized using a Uniform quantizationto reduce the number of bits and
to compress the image more efficiently. Then zigzag scan is used to scan the
less frequent coefficients. The quantized coefficients are encoded using any of
the encoding techniques. The reverse process of the compression phase is used
to decompress the image.

N and Sreeleja N Unnithan 4 state that the basic concept of image
compression, First the image is encoded by the encoder where the image is
converted into bit streams then the bit stream is decoded. This bit stream is
produced as an output image .When the output image has higher compression ratio
than the bit stream, and then image compression occurs. Different types of
redundancies are removed by assigning less number of bits to more frequent grey
levels and then assigning more number of bits to less frequent grey levels.
Various lossless compression techniques such as Huffman Encoding, Run Length
Encoding Arithmetic Coding, Entropy Encoding, Lempel–Ziv–Welch Coding and lossy
compression techniques such as Predictive coding and Transform coding are

Jindal 5 states the various techniques that can be applied to different
images. An image is a 2D array of pixels. A pixel is single point in a raster
image. An RGB pixel consists of three dots of light each corresponds to the
brightness of the particular color. A YUV color space consists of Ycbcr
components. Several lossy compression techniques are discussed. Transformation
coding is used to change the pixels in the original image into frequency domain
coefficients. Vector Quantization is a usually a block of pixel values. A given
image is then partitioned into non-overlapping blocks (vectors) called image vectors.
Fractal Coding the essential idea here is to decompose the image into different
segments by using different processing techniques such as color separation,
edge detection, and texture analysis. In Block Truncation Coding the image is
divided into non overlapping blocks of pixels. Various techniques are also
included in lossless compression such as Run Length Encoding. It replaces
sequences of identical pixels called runs by shorter symbols. Huffman Encoding
is a general technique for coding symbols based on their statistical occurrence
frequencies.LZW (Lempel-Ziv–Welch) is a dictionary based coding. Dictionary
based coding can be static or dynamic. Area coding is an enhanced form of run
length coding, reflecting the two dimensional character of images.

YAO, and Guizhong LIU 6 states that color image compression use contrast
sensitivity characteristics of human visual system. First the image is
converted from RGB color space to YUV color space. Then the image is divided
into three components such as luminance, red-green, Yellow-blue. Then each
component is divided into 8*8 block. The DCT transform is carried out for each
of the blocks in three components. Then, three quantization matrices based on
the CSF model are built, and the frequency spectrum coefficients of three
component images are selectively quantized using the matrices. Finally the
quantized coefficients are encoded using the Huffman Algorithm.




paper proposes the JPEG color image compression with image enhancement
employing Discrete Cosine Transform. The proposed scheme initially converts the
image   to YUV color space from an RGB
color Space as human eyes are sensitive to chrominance rather than luminance.
Then the converted image is enhanced and it is divided into 8*8 blocks. For
each block DCT (Discrete cosines transform) is applied. The coefficients
generated by the DCT are qunatized.Uniform Quantization is a technique where
the compression is done to a greater extent. Then Huffman coding is adopted to
encode the coefficients generated by the uniform quantization. Huffman Coding is lossless
compression algorithm that assigns variable length code to input characters.
The variable length code that is assigned to the input characters is called
prefix code. The code that is assigned to one character will not be the code
for some other character. If the input character appears more frequently they
are assigned as smallest code whereas if the input character appears least
frequently then it is assigned as largest code. Huffman Coding generates bit
stream. Thus the compressed image will be retrieved. Decompression is the
reverse process of compression. Various metrics such as the peak signal to Nose
ratio (PSNR) and Compression ratio (CR) is carried out to compare the
compressed image with the original image (7).








Proposed Method Taxonomy




proposed scheme first converts the input image  
from RGB color space   to YUV
color space as Fig.2

output of RGB to YUV color space


Then the image is enhanced using a histogram
equalization technique as shown in Fig.3


output of enhanced image


The enhanced image is compressed using a jpeg compression
algorithm after that DCT is applied on the enhanced image and IDCT is also
applied as
shown in Fig.4


output of DCT and IDCT



in the DCT image, quantization is applied which quantizes the image Then
dequantization is also applied to retain the DCT based image as shown in Fig.5





Output of quantization and dequantization


coding encodes and decodes the quantized image. Thus the decoded image is
converted to RGB color space from YUV color space as shown in Fig.6



Fig.6.   Output of decoded image




paper proposes the JPEG color image compression with image enhancement
employing Discrete Cosine Transform. Various performance metrics are calculated such as
signal to Noise Ratio) is used to find the ratio between the
maximum possible pixel value of an image and of corrupting noise that affects
the fidelity criteria of an image.MSE (Mean Square Error) is the cumulative
squared error between the compressed image and the Original image. It will
measure the quality of compressed image and Compression Ratio (CR) is used to
find the ratio between the original image and the corresponding compressed image
(7). Thus the results indicate that the proposed
scheme performs better than JPEG image compression algorithm.



Fig.7. Results of
performance analysis







paper discusses the enhanced image compression that is adopted for color
images. The aim is to obtain higher compression ratios and display images in
finer quality. Here the image is compressed by adopting techniques like DCT,
quantization and entropy coding. The inverse process is adopted to retrieve the
input image. We calculated different performance matrices such as PSNR and CR
for both JPEG compression and proposed method. Thus the results show the
proposed scheme attains finer image quality by increasing CR as well as PSNR of
compressed image.










1 A.H.M. Jaffar Iqbal
Barbhuiya, Tahera Akhtar Laskar, K. Hemachandran; 2014 Sixth International

      Computational Intelligence and
Communication Networks.

2A.M. Raid,
W.M.Khedr, M. A. El-dosuky and Wesam Ahmed; Jpeg Image Compression Using
Discrete Cosine Transform

    Survey; International Journal of Computer
Science & Engineering Survey (IJCSES) Vol.5, No.2, April 2014

3 Jaspreet Kaur,
Amandeep Kaur; Image Contrast Enhancement method based on Fuzzy Logic and
Histogram Equalization;

      International Research Journal of
Engineering and Technology Volume: 03 Issue: 05 |May-2016

4 Surabhi N, Sreeleja
N Unnithan; Image Compression Techniques: A Review; 2017 IJED | Volume 5, Issue

5 Ridhi Jindal; a
Review on Recent Development of Image Compression Techniques; International
Journal of Advance Research

      Development.Volume1, Issue1, 2017

6 Juncai YAO, and
Guizhong LIU; A Novel Color Image Compression Algorithm Using the Human Visual
Contrast Sensitivity

     Characteristics; PHOTONIC SENSORS, Vol. 7,
No.1, 2017: 72–81

7 Avudaiappan. T,
Ilam parith T, Balasubramanian. R, Sujatha K; Performance analysis on lossless image compression

     for general images;
Journal of Pure and Applied MathematicsVolume
117 No. 10 2017, 1-5

8 Sanjana C.Shekar,
D.J.Ravi; Image Enhancement and Compression using Edge Detection

     Technique; International Research Journal
of Engineering and Technology (IRJET); Volume:04 Issues: 05| May-2017