1 A JPEG-LS Based Lossless/Lossy Compression Method for Two-Dimensional Electrophoresis Images Source: 2003 International Conference on Informatics, Cybernetics, and Systems Authors: Kevin I-J Ho, Tung-Shou Chen, Hui-Fang Tsai, Mingli Hsieh, and Chia-Chun Wu Speaker: Chia-Chun Wu ( 吳吳吳 ) Date: 2004/12/09
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
A JPEG-LS Based Lossless/Lossy Compression Method
for Two-Dimensional Electrophoresis Images
Source: 2003 International Conference on Informatics, Cybernetics, and Systems
Authors: Kevin I-J Ho, Tung-Shou Chen, Hui-Fang Tsai, Mingli Hsieh, and Chia-Chun Wu
Speaker: Chia-Chun Wu (吳佳駿 )Date: 2004/12/09
2 NCHU
Outline
IntroductionSchemaCompression MethodDecompression MethodResultsConclusion
3 NCHU
We use Lossless and Near-Lossless compress the important areas and unimportant areas in Two-Dimensional Electrophoresis (2D-Gel) images .
Our system improves traditional JPEG-LS to enhancing the compressed image quality.
Introduction
4 NCHU
Schema (1/2)
Compression flow chart
3
Original 2D-Gel ImageOriginal 2D-Gel Image
Detect Protein’s AreasDetect Protein’s Areas
1
Record Boolean Value of Important Areas
Record Boolean Value of Important Areas
2
JPEG-LS Near-Lossless CompressJPEG-LS Near-Lossless Compress
５
4
Write Difference Value of Original Image’s Important Areas
Write Difference Value of Original Image’s Important Areas
Difference value Record FileDifference value Record File
６
Near-Lossless Compressed FileNear-Lossless Compressed File
５
5 NCHU
Decompression flow chart
Add Difference to Image’s Pixel Value
Add Difference to Image’s Pixel Value
Keep Important Information of 2D-Gel Image
Keep Important Information of 2D-Gel Image
JPEG-LS Near-Lossless Decompress
JPEG-LS Near-Lossless Decompress
３
Near-Lossless Compressed FileNear-Lossless Compressed File Difference value Record FileDifference value Record File
２
Near-Lossless Decompressed 2D-Gel Image
Near-Lossless Decompressed 2D-Gel Image
３ ４
Schema (2/2)
１
6 NCHU
Compression Method (1/5)
Original 2D-Gel image
－ This is an original 2D-Gel image. X-axis represented the PH value of protein and Y-axis represented the amount molecular weight.
Fig. 1 Original 2D-Gel image X-axis
Y-axis
73 72 75 78 71
71 75 18 4 74
76 5 15 16 10
73 18 23 28 74
75 74 10 73 77
7 NCHU
Compression Method (2/5)
Fetching protein’s area.
－ To collect the important protein ‘s areas of 2D-Gel image. The colourful areas will treat as important areas, and white areas will treat as unimportant areas.
Fig. 2 The important part of 2D-Gel image
0 0 0 0 0
0 0 18 4 0
0 5 15 16 10
0 18 23 28 0
0 0 10 0 0
8 NCHU
Compression Method (3/5)
Transform the important parts to Boolean value
－ Boolean value True(1) represents important areas, whereas False (0) represents unimportant areas.
Fig. 3 Boolean value record file of important part
0 0 0 0 0
0 0 1 1 0
0 1 1 1 1
0 1 1 1 0
0 0 1 0 0
9 NCHU
Compression Method (4/5)
Image after JPEG-LS Near-Lossless compression
－ This is an 2D-Gel image after traditional JPEG-LS Near-Lossless compression.
Fig. 4 Image after JPEG-LS Near-Lossless compression
71 73 77 76 74
72 74 17 6 73
79 3 18 13 8
73 21 23 26 74
74 77 12 75 76
10 NCHU
Compression Method (5/5)
Difference value records important part
－ The difference value of 2D-Gel image via the original image and lossless compression will store in a record file.
Fig.5 Difference value record file
0 0 0 0 0
0 0 1 -2 0
0 2 -3 3 2
0 -3 0 2 0
0 0 -2 0 0
11 NCHU
Compression Example
73 72 75 78 71
71 75 18 4 74
76 5 15 16 10
73 18 23 28 74
75 74 10 73 77
Original 2D-Gel Image
71 73 77 76 74
72 74 17 6 73
79 3 18 13 8
73 21 23 26 74
74 77 12 75 76
Image after JPEG-LS Near-Lossless compression
0 0 0 0 0
0 0 1 -2 0
0 2 -3 3 2
0 -3 0 2 0
0 0 -2 0 0
Difference value record file
- =
12 NCHU
Decompression Method (1/3)
Fig 6.Image after JPEG-LS Near-Lossless compression
The image of Near-Lossless decompression
－ This is a decompressed image after traditional JPEG-LS Near-Lossless compression.
71 73 77 76 74
72 74 17 6 73
79 3 18 13 8
73 21 23 26 74
74 77 12 75 76
13 NCHU
Decompression Method (2/3)
Modify the protein’s area of important part
－ Next, we base on the difference value of pixels for modifying the protein’s area of important parts.
Fig. 7 Difference value record file
0 0 0 0 0
0 0 1 -2 0
0 2 -3 3 2
0 -3 0 2 0
0 0 -2 0 0
14 NCHU
Decompression Method (3/3)
Fig. 8 Our system’s lossless compression of important part.
The lossless image of our system’s important areas
－ This complete 2D-Gel image is though traditional JPEG-LS Near-Lossless compression technique.
71 73 77 76 74
72 74 18 4 73
79 5 15 16 10
73 18 23 28 74
74 77 10 75 76
15 NCHU
Decompression Example
Image after JPEG-LS Near-Lossless compression
Difference value record file
Our system’s lossless compression of important part
+ =
71 73 77 76 74
72 74 17 6 73
79 3 18 13 8
73 21 23 26 74
74 77 12 75 76
0 0 0 0 0
0 0 1 -2 0
0 2 -3 3 2
0 -3 0 2 0
0 0 -2 0 0
71 73 77 76 74
72 74 18 4 73
79 5 15 16 10
73 18 23 28 74
74 77 10 75 76
16 NCHU
Results (1/4)
Fig. 9 Partial magnify image of original 2D-Gel image
Fig. 9 is the result of the amplification of dotted frame in Fig. 1.
17 NCHUFig. 10 Partial magnify image of traditional JPEG-
LS.
Results (2/4)
Fig.10 is the result of the amplification of dotted frame in Fig. 4.
18 NCHUFig. 11 Partial magnify image of our system.
Results (3/4)
Fig. 11 is the result of the amplification of dotted frame in Fig. 8.
19 NCHU
Result (4/4)Table 1. The Comparison of image quality in traditional JPEG-LS
Near-Lossless with our system (PSNR value). Unit:dB
Results (4/4)
34.57 35.69 36.42 37.47 38.72 40.13 41.85 43.89 46.69 51.08 Our System
33.50 34.56 35.33 36.40 37.65 39.04 40.77 42.82 45.66 50.10 Jpeg-ls2DGel09
34.43 35.27 36.13 37.17 38.31 39.63 41.21 43.16 45.90 50.48 Our System
33.79 34.64 35.50 36.54 37.69 39.01 40.61 42.59 45.35 49.94 Jpeg-ls2DGel08
34.63 35.45 36.35 37.36 38.44 39.77 41.36 43.32 46.03 50.60 Our System
33.83 34.64 35.54 36.57 37.66 39.01 40.61 42.58 45.34 49.93 Jpeg-ls2DGel07
35.70 36.15 37.12 38.14 39.45 40.75 42.39 44.11 46.90 51.34 Our System
34.19 34.77 35.73 36.74 38.03 39.37 41.03 42.78 45.68 50.14 Jpeg-ls2DGel06
34.71 35.48 36.41 37.41 38.57 39.89 41.34 43.36 45.82 51.16 Our System
33.51 34.31 35.24 36.23 37.41 38.75 40.25 42.30 44.85 50.08 Jpeg-ls2DGel05
37.44 32.63 38.48 39.55 40.54 42.00 41.16 41.73 45.58 50.98 Our System
36.11 32.16 37.32 38.37 39.41 40.90 40.47 41.21 44.98 50.29 Jpeg-ls2DGel04
35.81 36.27 36.90 37.59 38.78 40.63 42.31 44.13 46.96 51.29 Our System
34.42 34.97 35.61 36.41 37.57 39.43 41.07 42.93 45.81 50.23 Jpeg-ls2DGel03
35.88 36.75 37.81 38.57 39.80 41.15 42.71 44.70 47.35 51.63 Our System
33.85 34.73 35.74 36.61 37.83 39.20 40.82 42.85 45.60 50.02 Jpeg-ls2DGel02
34.5335.0236.3037.3238.3539.7141.3643.3946.1050.68Our System
33.7534.2335.5336.5537.5738.9740.6342.6745.3750.00Jpeg-ls2DGel01
109876
54321Lossless
lever
34.57 35.69 36.42 37.47 38.72 40.13 41.85 43.89 46.69 51.08 Our System
33.50 34.56 35.33 36.40 37.65 39.04 40.77 42.82 45.66 50.10 Jpeg-ls2DGel09
34.43 35.27 36.13 37.17 38.31 39.63 41.21 43.16 45.90 50.48 Our System
33.79 34.64 35.50 36.54 37.69 39.01 40.61 42.59 45.35 49.94 Jpeg-ls2DGel08
34.63 35.45 36.35 37.36 38.44 39.77 41.36 43.32 46.03 50.60 Our System
33.83 34.64 35.54 36.57 37.66 39.01 40.61 42.58 45.34 49.93 Jpeg-ls2DGel07
35.70 36.15 37.12 38.14 39.45 40.75 42.39 44.11 46.90 51.34 Our System
34.19 34.77 35.73 36.74 38.03 39.37 41.03 42.78 45.68 50.14 Jpeg-ls2DGel06
34.71 35.48 36.41 37.41 38.57 39.89 41.34 43.36 45.82 51.16 Our System
33.51 34.31 35.24 36.23 37.41 38.75 40.25 42.30 44.85 50.08 Jpeg-ls2DGel05
37.44 32.63 38.48 39.55 40.54 42.00 41.16 41.73 45.58 50.98 Our System
36.11 32.16 37.32 38.37 39.41 40.90 40.47 41.21 44.98 50.29 Jpeg-ls2DGel04
35.81 36.27 36.90 37.59 38.78 40.63 42.31 44.13 46.96 51.29 Our System
34.42 34.97 35.61 36.41 37.57 39.43 41.07 42.93 45.81 50.23 Jpeg-ls2DGel03
35.88 36.75 37.81 38.57 39.80 41.15 42.71 44.70 47.35 51.63 Our System
33.85 34.73 35.74 36.61 37.83 39.20 40.82 42.85 45.60 50.02 Jpeg-ls2DGel02
34.5335.0236.3037.3238.3539.7141.3643.3946.1050.68Our System
33.7534.2335.5336.5537.5738.9740.6342.6745.3750.00Jpeg-ls2DGel01
109876
54321Lossless
lever
20 NCHU
Conclusion
We store the unimportant areas by Near-Lossless method. But we store important areas by Lossless method. It is very important to medical images.
Under different lossless level, we can find out our system has better image quality than traditional JPEG-LS.
Therefore, how to compress the size of record file and detect the protein’s location more correctly becoming an important topic in the future.
Related Documents
