HIGH PERFORMANCE CAR AUDIO
MODEL:CR-CHANNEL ELECTRONIC CROSSOVER NETWORK SYSTEM

www.pyramidcaraudio.com

INTRODUCTION
This CR-82 is an Electronic Crossover Network with which you can install various types of multi-channel amplifier systems in a car. This CR-82 is 8-channel crossover system (Subwoofer, Mid-range, High-pass and Flat). Before designing and installing your system. Please read this manual carefully so that you can set up the system that practically suits you. Getting the CR-82 in the center of your multi-amplifier car stereo system. Your car stereo system will provide you with the ultimate sound reproduction.
FEATURES & SPECIFICATIONS
2-Way stereo input. Subwoofer/Front 4-Way stereo output. Subwoofer/Mid-range/High Pass/Flat DC to DC converter power supply. Individual output level controls for each stereo channel. Variable subwoofer boost 0- 10dB at 40Hz. Subwoofer channel 45/80/150/200Hz. Subwoofer phase shift 0-180 degree. Subwoofer STEREO/MONO switch. Subwoofer input ON/OFF switch. Selectable band pass channel 100/150/200/Flat to 2.5k/3.5k/5k/7k Signal to Noise ratio :>90dB. Distortion : less than 0.02% Power supply :12-16V DC, Neg. Gnd.

FUNCTIONS

1 CROSSOVER POINT SELECTOR (SUBWOOFER) Controls the subwoofer output frequency at 45,80,150,200Hz 2 SUBWOOFER INPUT SWITCH When your car radio has subwoofer output to connect the unit's subwoofer input;Push the switch to the "OFF" Position. If your car radio only has line output to connect the unit 's line input; Push the switch to the "ON" Position
3 PHASE SHIFTER SELECTOR (SUBWOOFER) Allows you to change the phase of your subwoofer from 0 to 180 degrees to help compensate for timing differences between drivers. 4 BASS BOOST CONTROL (40Hz) Frequencies around 40Hz can be boosted at the maximum of +10dB. 5 STEREO/MONO SELECTOR (SUBWOOFER) Set the selector to MONO to make subwoofer output monaural. 6 FLAT CHANNEL GAIN CONTROL By turning this control clockwise you increase flat channel level. 7 SUBWOOFER GAIN CONTROL By turning this control clockwise you increase subwoofer channel level. 8 MIDDLE RANGE CROSSOVER POINT SELECTOR The high Pass frequency(100,150,200Hz,Flat) should match the low crossover point, the low frequency(2.5k,3.5k,5k,Flat)should match the high crossover point. 9 MIDDLE RANGE GAIN CONTROL By turning this control clockwise you increase middle range channel level. 0 PHASE SHIFTER SELECTOR(HIGH PASS) Allows you to change the phase of your high pass from 0 to 180 degrees to help compensate for timing differences between drivers. q HIGH PASS CROSSOVER POINT SELECTOR Control the high pass output at 2.5k,3.5k,5k,7k. w HIGH PASS GAIN CONTROL By turning this control clockwise you increase high pass channel e POWER ON INDICATOR The LED light up when the units is powered on.
r SUBWOOFER INPUT CONNECTORS t FRONT INPUT CONNECTORS y FLAT OUTPUT CONNECTORS u SUBWOOFER OUTPUT CONNECTORS i MIDDLE RANGE OUTPUT CONNECTORS o HIGH PASS OUTPUT CONNECTORS
ELECTRONIC CONNECTIONS & WIRING

Power connection

B + ( l 2 V ) :Connect a red wire(at least l0-12 gauge)to the car battery or other power source. REMOTE :Connect an orange wire to remote activating(l2V DC)wire of car stereo or equalizer. GND :Connect a black wire(at least l0-12 gauge)to the car chassis for ground connection.

Signal connection

TROUBLE SHOOTING SECTION
1. SIGNAL PROCESSOR DOES NOT TURN ON: Check all fuses. Check all power(positive ,remote) wire connections. Check that ground wire is properly connected. 2. LED POWER ON INDICATOR NOT GOING ON: Same as sign l remedy. 3. SOUND DISTORTION AT LOW VOLUME LEVEL: Output level not set correctly. 4. LEVEL OF SOUND IS LOW: Check input gain control ,match to deck output. Check the RCA patch cords for loose or misconnected cables. 5. A WHINING SOUND CAN BE HEARD THROUGH THE SPEAKERS AT LOW VOLUME LEVEL WITH RUNNING ENGINE: Check the power wire(red)with good connection directly to the battery and ground point must make good contact with chassis ground.

INSTALLATION

New Approach for Efficient Archiving of Electronic Documents
R&D Group: Roumen Kountchev Vladimir Todorov

Outlines

1.Main 1. Main problems with documents archiving 2. Adaptive image compression 3. Lossless IDP compression 4. Experimental Results and Comparison with JPEG and JPEG 2000 5.Conclusion 5. Conclusion
Documents Archiving Main problems
1. Documents to be archived: text documents (newspapers, magazines, books), photos, medical images, etc. 2. Main methods for image compression consist of two large groups: statistical and psycho-visual. Statistical methods:
Run-Length Coding (RLC); Lempel-Ziv-Welch (LZW) coding, based on dictionaries; Huffman coding; Arithmetic coding; Linear prediction coding. Psycho-visual methods: based on Linear orthogonal transforms: Karhunen-Loeve Transform (KLT), Discrete Cosine Transform (DCT), Walsh-Hadamard Transform (WHT), wavelet transforms, etc., Vector quantization (VQ), decomposition, fractal transforms, etc. The traditional compression techniques are usually based on the standards for image compression, JPEG or JPEG2000; ISO/IEC JPEG-LS Standard.
Disadvantages of the JPEG and JPEG 2000 standards

The original test image

The test image after JPEG compression Compression ratio (CR) = 15
Adaptive image compression
The new approach for documents archiving is based on the image analysis and is done in accordance with the results obtained:
The parts of the image, which contain pictures, are compressed using
Inverse Pyramidal Decomposition; The parts, containing text are losslessly compressed. The classification is based on the image histogram analysis. 5
Block diagram of the Inverse Difference Pyramid (IDP) coder

[E1] 1

4 [E1 ]

[E13] 1

[E16] 1

Truncated OT n 2n 2

~1 [S2 ]

~4 [S2 ]

~16 [S2 ]

~13 [S2 ]

~ (u, v) scoeff.

~ [E ]

~ ~ [E ] [E ] ~ ~ [E ] [E ]

Truncated OT

Inverse OT 2n 1 2n 1

[E ] [E ]

~1 ~ [S1 ] [S12 ]

2n 1 2n 1

~ ~ [S13 ] [S14 ]

~ [B0 ]

Inverse OT

Truncated OT 2n 2n

~ [S0 ]

Modified Huffman coding

Adaptive RLE

Sub-band arrangement

Image compression with IDP
Spectral space Layer p=coeff. Pixel space
Sub-block 4x4 PSNR=32.1 dB BR=1.88 bpp

IDCT-DCT

~ ~ [B0] [E0] [E1]
Sub-block 8x8 PSNR=27.1 dB BR=0.46 bpp
IDCT-DCT Layer p=coeff. Layer p=coeff.

3-layer Spectral Pyramid

~ [S1 ]

~ [E0 ]

~ ~ [B0] [E0]
Block 16x16 PSNR=23.7 dB BR=0.11 bpp

~ [B0]

Image Restoration
Three-layer IDP representation of the image Lena (512512 pixels, 8 bpp)

Image Quality Evaluation

The quality of the restored image is evaluated by its PSNR (Peak Signal/Noise Ratio), compared with the original.

Original Image Lena 512x512, 8 bpp
Restored Image after IDP compression PSNR=32.16 dB, CR=16.67 (0.48 bpp) 8
Original Image Sportsman 1940x2004, 24 bpp
Restored Image after CRIDP=400 (0.06 bpp)

Comparison IDP and JPEG

CR=400 (IDP)

CR=400 (JPEG-DCT)

Comparison of the computational complexity of the IDP and DWT- 5/3 tap (JPEG 2000)

Operation

Decomposition
for one pixel in the coder
for one pixel in the decoder
M for one pixel in the coder
M for one pixel in the decoder

IDP DWT

(JPEG2000)

14.31 31.5

7.15 36.75

2.18 21

1.09 21
- number of additions; M number of multiplications.
Method for Lossless Image Compression
Main application areas: Fingerprints Texts Graphics Contours Monochrome images Images with slowly changing color or brightness Computer games and cartoons
Lossless image compression - results
Test-image Crosses, 256x256, 8 bpp

CR=40 (IDP) (PSNR = )

CR=40 (JPEG2000) (PSNR = 24 dB)
Test-image Squares, 256x256, 8 bpp

CR=468 (RSPP) (PSNR = )

CR=468 (JPEG2000)

Fingerprints

8 bpp, 288x353 pixels

Image 01

Image 02

Image 03

Results -IDP and JPEG 2000
L o s s l e s s i m a g e c o m p r e s s i o n - fi n g e r p r i n t s (g r a y sc a le , 8 b p p )

Image No ID P

C o m p res sio n ra t io

JPEG 2000

Adaptive compression with IDP decomposition

Test Compound image

The histograms of the picture and text parts
On the basis of the histogram analysis are fixed the two regions of interest in the processed image

Experimental results

Test image Formula (980 x 560 pixels, grayscale)
JPEG 2000 lossless IDP Lossless

8,35 28,31

19 400
Comparison with the JPEG 2000 standard

Example graphic images

Circles

Squares

Watermark 1

Watermark 2 19

Compression Ratios, obtained for graphic test images with ARL No. Image 3 Circles Crosses Squares IDP Lossless 9,46 10,05 61,59 JPEG2000 8,57 7,45 43.92 Image size 256x256 256x256 256x256

SQ Wmark1

69,28 97,11

60,56 57,32

256x256 512x512
Test image Squares, restored after lossless IDP: CR=770 JPEG2000 lossless CR = 44 IDP is 17,5 times more efficient
Test image Sq1, restored after lossless IDP: CR=261 JPEG200 lossless CR = 61 IDP is 4,27 times more efficient 21
Test image Slope, after lossless IDP, CR = 73,4 Lossless JPEG200 CR = 22,1 IDP is 3,3 times more efficient
Test image WM after lossless IDP, CR= 82 Lossless JPEG200 CR = 17,3 IDP is 4,74 times more efficient

Comparison results

Text test images

Compression ratio

15 IDP JPG2000 10

Im age No

IDP decomposition for multispectral images Image IPD 2lvl [KB] 124 Lvl1 [KB] 33,1 24,0 36,7 90,0 274,7 15,26 Total [KB] CR PSNR [dB] 32,1 45,0 36,0 31,0

Red Blue Pan Pir

JPEG compression Image Red Blue Pan Pir JPEG size [KB] Total [KB] CR PSNR [dB] 32,25 31,3 31,3 32,2 24
Digital watermarking with IDP

Original image Peppers

Watermark image
Extracted fragile watermark

Watermarked test image

Original image Lena
Watermarked image with phase angle in the imaginary part of the Complex Hadamard coefficients, = 12
Example for resistant watermarking

PSNR, [dB]

20 Watermark depth, [deg]
The PSNR of the watermarked image in a function of the angle

Conclusions

Main advantages of the new method:
1. It permits the recognition of texts and pictures in compound images and their compression with most suitable algorithms, setting corresponding ROIs; It permits the processed pictures to be transferred layer by layer, with increasing quality; The computational complexity of the new method is lower than that of the JPEG 2000 standard; It is very efficient for compression of still images comprising texts, graphics and pictures; It offers efficient image compression, suitable for creation of distancelearning courses, accessed via Internet. 27
The IDP decomposition permits easy watermark insertion in the consecutive image layers with increasing resolution. The resistant watermark data is added to the imaginary part of the coded image data and does not influence the restored image quality. The software implementation in C++ (Windows environment) of the presented approach proved the method efficiency. The main advantages of the presented algorithm for resistant watermarking are: - The ability for blind watermark detection (without using the original image); - The watermark transparency; - The resistance of the inserted watermark against various attacks, such as compression and highfrequency filtration; -The high efficiency more than 100 bits for halftone images of size pixels with retained visual quality. The most important applications are for content protection of any kind of documents, medical images, etc., i.e. all cases, when the image quality should not be changed and the confidential information is preserved. The new method offers significant resources for efficient content protection of images in large databases. The main advantages of the method for fragile watermarking are: The watermark is inserted as an additional decomposition layer and does not influence the protected image quality; Any change in the extracted watermark evidences unauthorized access and image editing; The knowledge of the algorithm and the possession of the decoding tools do not permit the watermark extraction without having the password. The ability to insert more than one watermark in the same image ensures higher security for the original image content and permits the creation of information systems with hierarchical access control. The so developed methods for multiple watermarking offer wide abilities for implementation in various application areas.

References

1. 2. R. Kountchev, V. Haese-Coat, J. Ronsin. Inverse Pyramidal Decomposition with multiple DCT. Elsevier, Signal Processing: Image Communication, V. 17/ 2, Feb. 2002, pp. 201-218. R. Kountchev, M, Milanova, C. Ford, R. Kountcheva. Multi-layer Image Transmission with Inverse Pyramidal Decomposition. Book chapter in: Computational Intelligence for Modelling and Predictions, S. Halgamuge, L. Wang (Eds.), Vol. 2, Ch. 13, Springer-Verlag, 2005, pp. 179-196. R. Kountchev, R. Kountcheva. Image Representation with Reduced Spectrum Pyramid. Book Chapter in: New Directions in Intelligent Interactive Multimedia, Eds. G. Tsihrintzis et al, Springer Verlag, 2008, pp. 275-284. 3.
4. R. Kountchev, Vl. Todorov, R. Kountcheva. Flexible Archiving of Old Handwritten Documents. Proc. of the WSEAS Intern. Conf. on Signal Processing (SIP08), Istanbul, Turkey, May 27-29, 2008, pp. 13-18. 5. R. Kountchev, Vl. Todorov, R. Kountcheva. Adaptive preprocessing of scanned documents. Proc. of the WSEAS Intern. Conf. on Mathematical Methods and Computational Techniques in Electrical Engineering (MMACTEE08), Sofia, Bulgaria, May 2-4, 2008, pp. 34-39. 6. R. Kountchev, Vl.Todorov, R. Kountcheva. Fragile and Resistant Image Watermarking Based on Inverse Difference Pyramid Decomposition. WSEAS Transactions on Signal Processing. Issue 3, Volume 6, July 2010, pp. 101-112. R. Kountchev, Vl.Todorov, R. Kountcheva. New Method for Lossless Data Compression Based on Adaptive Run-Length Coding. Book chapter in: Advanced Computational Technologies, C. Enachescu, F. Filip, B. Iantovics (eds.). Romanian Academy Publishing House. (In press). R. Kountchev, R. Kountcheva. Comparison of the structures of the Inverse Difference and Laplacian Pyramids for Image Decomposition. Proc. of the 45th Intern. Scientific Conf. on Information, Communication and Energy Systems and Technologies (ICEST10), June 2010, Macedonia, pp. 33-36.