US2004101193A1PendingUtilityA1

Document analysis method to detect BW/color areas and corresponding scanning device

Priority: Sep 22, 1998Filed: Jun 25, 2003Published: May 27, 2004
Est. expirySep 22, 2018(expired)· nominal 20-yr term from priority
H04N 1/56H04N 1/642
44
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

The invention relates to a document analysis method to detect BW/color areas comprising the following steps: getting an input image data to be split in elementary information; analysing the input data to get the colorfulness of each elementary information; creating a first and a second output layers corresponding to the BW and color elementary information respectively; applying a first set of atomic operations to the BW layer and a second different set of atomic operations to the color layer; and combining the BW and color layers in order to obtain a desired format for the output data. The invention also relates to a scanning device to acquire documents including an input that receives an input image data, and is connected to a layer creator component, which in turn outputs a first and a second layer, such layers having different data compression rate. Finally, the invention relates to a method for acquiring a document based on the analysis of the content of the document itself.

Claims

exact text as granted — not AI-modified
1 . A document analysis method to detect BW and color areas comprising the following steps: 
 step 1) getting an input image data to be split into elementary unit of information;    step 2) analysing the input data to get a colorfulness indicator of each elementary unit of information of the input data;    step 3) assigning each elementary unit of information to either a BW layer or a color layer depending on the colorfulness indicator of said each elementary unit of information;    step 4) determining, in order to obtain a desired format for the output data, either to select or to combine the BW and color layers.    
     
     
         2 . A document analysis method according to  claim 1 , characterized in that the step of analysing the input. data further comprises the step of getting the input data into a Chroma space format.  
     
     
         3 . A document analysis method according to  claim 1 , characterized in that it further comprises the step of converting the input data from the RGB format to a Chroma space format.  
     
     
         4 . A document analysis method according to  claim 2 , characterized in that it further comprises the step of down sampling a chroma indication channel.  
     
     
         5 . A document analysis method according to  claim 4 , characterized in that it further comprises the step of applying a threshold or a rule for the colorfulness indicator to the down sampled data.  
     
     
         6 . A document analysis method according to  claim 5 , characterized in that it further comprises the step of labeling ON the elementary information having a colorfulness above the threshold and OFF the elementary information having a colorfulness lower than the threshold.  
     
     
         7 . A document analysis method according to  claim 6 , characterized in that the threshold value depends on the final device which receive the image data.  
     
     
         8 . A document analysis method according to  claim 7 , characterized in that the BW elementary information are employed to assemble a first layer (TEXT) containing the portion of text comprised in the input data and the color elementary information are employed to assemble a second layer (IMAGE) containing the portion of image comprised in the input data.  
     
     
         9 . A scanning device to acquire documents characterized by comprising an input that receives an input image data, and is connected to a layer creator component, which in turn outputs a first and a second layer, such layers having different data compression rates.  
     
     
         10 . A scanning device according to  claim 9 , characterized in that said first layer assembles BW data and said second layer assembles color data.  
     
     
         11 . A scanning device according to  claim 10 , characterized in that said layer creator component comprises a threshold block connected in series to a compressor.  
     
     
         12 . A scanning device according to  claim 11 , characterized in that said compressor is a G 4  compressor.  
     
     
         13 . A scanning device according to  claim 12 , characterized in that said layer creator component further comprises a RGB to HLN converter, having an output channel connected to a series of a down scale device, an histogram and threshold selector, a look-up-table, a blob analysis block, a fill regions block and a further compressor.  
     
     
         14 . A scanning device according to  claim 13 , characterized in that said further compressor is a JPEG compressor.  
     
     
         15 . Method for acquiring a document based on the analysis of the content of the document itself, comprising the following steps: 
 getting an input image data;    creating a first layer containing the image information in a color format;    creating a second layer containing the image information in a BW format;    managing the first and the second layers in order to obtain a desired format for an output document.    
     
     
         16 . Method for acquiring a document as claimed in  claim 15 , characterized in that the image information contained in the first layer has a resolution lower that the image information contained in the first layer.  
     
     
         17 . Method for acquiring a document as claimed in  claim 16 , characterized in that the sum of a size of the first layer plus a size of the second layer is lower than a size of said first layer at the second layer higher resolution.  
     
     
         18 . Method for acquiring a document as claimed in  claim 15 , characterized in that the second layer is used for managing text information.

Join the waitlist — get patent alerts

Track US2004101193A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.