Printing quality enhancement via graphic/text detection method in compression (JPEG) image

• US 20050271284A1
• Filed: 06/07/2004
• Published: 12/08/2005
• Est. Priority Date: 06/07/2004
• Status: Active Grant

• Current Assignee: Lite-On Technology Corporation
• Original Assignee: Lite-On Technology Corporation
• Inventors: Chen, Chen-Chung, Chiang, Cheng-Huei
• Granted Patent: US 7,362,474 B2
• Time in Patent Office: Days
• US Class Current: 382/233
• CPC Class Codes: H04N 1/4053 with threshold modulated relative to input image data or vice versa
• CPC Class Codes: H04N 1/642 Adapting to different types of images, e.g. characters, graphs, black and white image portions
• CPC Class Codes: H04N 19/12 Selection from among a plurality of transforms or standards, e.g. selection between discrete cosine transform [DCT] and sub-band transform or selection between H.263 and H.264
• CPC Class Codes: H04N 19/122 Selection of transform size, e.g. 8x8 or 2x4x8 DCT; Selection of sub-band transforms of varying structure or type
• CPC Class Codes: H04N 19/126 Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
• CPC Class Codes: H04N 19/136 Incoming video signal characteristics or properties
• CPC Class Codes: H04N 19/196 being specially adapted for the computation of encoding parameters, e.g. by averaging previously computed encoding parameters processing of motion vectors H04N19/513
• CPC Class Codes: H04N 19/60 using transform coding

Abstract

The present invention relates to a printing quality enhancement via graphic/text detection method in compression (JPEG) image, wherein an unit of a n×n pixels of plural compressed subimages of a compressed image is taken as a minimum feature extraction unit, and the subimages are taken as input unit, and after a feature attribute and non-feature attribute processing, an output unit is formed and printed.

Claims

1. A printing quality enhancement via graphic/text detection method in compression (JPEG) image, wherein an unit of n×

n pixels of plural compressed subimages of a compressed image is taken as a minimum feature extraction unit, and the subimages are taken as input unit, then an output unit is formed via a feature attribute and non-feature attribute processing, and finally the output unit is printed, such a method includes the following steps;

subimage decompression serving to transform the compressed subimage pixels from frequency domain into space domain, including a DCT;

GTD analysis serving to determine the attribute of the subimages by calculating detail features of the subimages based on the frequency parameter of DCT of the subimages;

weight adjustment serving to process the respective subimages that have been processed by the GTD analysis in a manner that respectively processes, based on different weight processing, the subimages having feature attribute and the subimages having non-feature attribute, and result of the weight processing is used to determine edge direction of the respective subimages;

output processing serving to smoothly distribute ink dots of the respective subimages of non-edge feature and to sharply distribute ink dots of the subimages of edge feature with halftoning error diffusion.

2. The printing quality enhancement via graphic/text detection method in compression (JPEG) image as claimed in claim 1, wherein the GTD analysis includes a comparative program which serves to compare the high frequency ratio H(f) of the respective subimages with a preset reference value T(t), the subimages have text (edge) feature if Hx_ij (f)>

T(t), and the subimages have graphic (non-edge) feature if Hx_ij (f)<

T(t), a method for calculating the high frequency ratio is explained as follows;

${\mathrm{Hx}}_{\mathrm{ij}}<br><br>\left(f\right)=\frac{\left[\underset{p=0}{\overset{7}{\sum <br><br>}}<br><br>\underset{q=0}{\overset{7}{\sum <br><br>}}<br><br><br><br>{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(p,q\right)<br><br>-{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(0,0\right)\right]}{\underset{p=0}{\overset{7}{\sum <br><br>}}<br><br>\underset{q=0}{\overset{7}{\sum <br><br>}}<br><br><br><br>{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(p,q\right)<br><br>}$

where the T(t) is a preset critical value.

3. The printing quality enhancement via graphic/text detection method in compression (JPEG) image as claimed in claim 1, wherein the weight adjustment includes a detail feature weight WHx_ij (f), a calculation method for WHx_ij (f) is as follows:

${\mathrm{WHx}}_{\mathrm{ij}}<br><br>\left(f\right)=\frac{\left[\underset{p=0}{\overset{7}{\sum <br><br>}}<br><br>\underset{q=0}{\overset{7}{\sum <br><br>}}<br><br><br><br>{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(p,q\right)<br><br>-{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(0,0\right)\right]}{\underset{p=0}{\overset{7}{\sum <br><br>}}<br><br>\underset{q=0}{\overset{7}{\sum <br><br>}}<br><br><br><br>{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(p,q\right)<br><br>}\times <br><br>\mathrm{Wi}$

where the Wi is a preset reference value.

4. The printing quality enhancement via graphic/text detection method in compression (JPEG) image as claimed in claim 1, wherein the weight adjustment further includes a directional weight DX_ij, and a calculation for the directional weight DX_ij is shown as follows:

Dx_ij;

If vx_ij (f)>

hx_ij then Dx_ij=vx_ij (f)×

Wi else Dx_ij=hx_ij(f)×

Wi

where

${\mathrm{vx}}_{\mathrm{ij}}<br><br>\left(f\right)=\max <br><br>\left\{\frac{\underset{p=1}{\overset{7}{\sum <br><br>}}<br><br><br><br>{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(p,q\right)<br><br>}{\underset{p=0}{\overset{7}{\sum <br><br>}}<br><br>\underset{q=0}{\overset{7}{\sum <br><br>}}<br><br><br><br>{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(p,q\right)<br><br>}❘<br><br>q=0,2,3,\mathrm{\dots <br><br>}<br><br>\text{\hspace{1em}<br><br>},7\right\}-\min <br><br>\left\{\frac{\underset{p=1}{\overset{7}{\sum <br><br>}}<br><br><br><br>{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(p,q\right)<br><br>}{\underset{p=0}{\overset{7}{\sum <br><br>}}<br><br>\underset{q=0}{\overset{7}{\sum <br><br>}}<br><br><br><br>{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(p,q\right)<br><br>}❘<br><br>q=0,2,3,\mathrm{\dots <br><br>}<br><br>\text{\hspace{1em}<br><br>},7\right\}$ ${\mathrm{hx}}_{\mathrm{ij}}<br><br>\left(f\right)=\max <br><br>\left\{\frac{\underset{q=1}{\overset{7}{\sum <br><br>}}<br><br><br><br>{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(p,q\right)<br><br>}{\underset{p=0}{\overset{7}{\sum <br><br>}}<br><br>\underset{q=0}{\overset{7}{\sum <br><br>}}<br><br><br><br>{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(p,q\right)<br><br>}❘<br><br>p=0,2,3,\mathrm{\dots <br><br>}<br><br>\text{\hspace{1em}<br><br>},7\right\}-\min <br><br>\left\{\frac{\underset{q=1}{\overset{7}{\sum <br><br>}}<br><br><br><br>{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(p,q\right)<br><br>}{\underset{p=0}{\overset{7}{\sum <br><br>}}<br><br>\underset{q=0}{\overset{7}{\sum <br><br>}}<br><br><br><br>{\mathrm{fx}}_{\mathrm{ij}}<br><br>\left(p,q\right)<br><br>}❘<br><br>p=0,2,3,\mathrm{\dots <br><br>}<br><br>\text{\hspace{1em}<br><br>},7\right\}$

the Wi is a preset reference value.

5. The printing quality enhancement via graphic/text detection method in compression (JPEG) image as claimed in claim 1, wherein the GTD analysis further includes a process for determination of feature of neighboring tiles, so as to improve correctness of the attribute of the respective subimages.

6. The printing quality enhancement via graphic/text detection method in compression (JPEG) image as claimed in claim 1, wherein a quantizer and an error filter are disposed between the input unit and the output unit and form a closed-loop auto-control system;

further, the weight adjustment further includes a detail feature weight and a directional weight, the detail feature weight is added to the quantizer and the directional weight is added to the error filter.

7. The printing quality enhancement via graphic/text detection method in compression (JPEG) image as claimed in claim 1, wherein the n×

n pixels of the respective subimages refer to n=8.