1GOCR(1) User's Manual GOCR(1)
2
6 gocr - command line text recognition tool
7
9 gocr [OPTION] [-i] pnm-file
10
12 gocr is an optical character recognition program that can be used from
13 the command line. It takes input in PNM, PGM, PBM, PPM, or PCX format,
14 and writes recognized text to stdout. If the pnm file is a single
15 dash, PNM data is read from stdin. If gzip, bzip2 and netpbm-progs are
16 installed and your system supports popen(3) also pnm.gz, pnm.bz2, png,
17 jpg, jpeg, tiff, gif, bmp, ps (only single pages) and eps are supported
18 as input files (not as input stream), where pnm can be replaced by one
19 of ppm, pgm and pbm.
20
22 -h show usage information
23 -i file
25 read input from file (or stdin if file is a single dash)
26 -o file
28 send output to file instead of stdout
29 -e file
31 send errors to file instead of stderr or to stdout if file is a
32 dash
33 -x file
35 progress output to file (file can be a file name, a fifo name or
36 a file descriptor 1...255), this is useful for GUI developpers
37 to show the OCR progress, the file descriptor argument is only
38 available, if compiled with __USE_POSIX defined
39 -p path
41 database path, a final slash must be included, default is ./db/,
42 this path will be populated with images of learned characters
43 -f format
45 output format of the recognized text (ISO8859_1 TeX HTML XML
46 UTF8 ASCII), XML will also output position and probability data
47 -l level
49 set grey level to level (0<160<=255, default: 0 for autodetect),
50 darker pixels belong to characters, brighter pixels are inter‐
51 preted as background of the input image
52 -d size
54 set dust size in pixels (clusters smaller than this are
55 removed), 0 means no clusters are removed, the default is -1 for
56 auto detection
57 -s num set spacewidth between words in units of dots (default: 0 for
59 autodetect), wider widths are interpreted as word spaces,
60 smaller as character spaces
61 -v verbosity
63 be verbose to stderr; verbosity is a bitfield
64 -c string
66 only verbose output of characters from string to stderr, more
67 output is generated for all characters within the string, the
68 underscore stands for unknown chars, this function is usefull to
69 limit debug information to the necessary one
70 -C string
72 only recognise characters from string, this is a filter function
73 in cases where the interest is only to a part of the character
74 alphabet, you can use 0-9 or a-z to specify ranges, use -- to
75 detect the minus sign
76 -a certainty
78 set value for certainty of recognition (0..100; default: 95),
79 characters with a higher certainty are accepted, characters with
80 a lower certainty are treated as unknown (not recognized); set
81 higher values, if you want to have only more certain recognized
82 characters
83 -u string
85 output this string for every unrecognized character (default is
86 "_")
87 -m mode
89 set oprational mode; mode is a bitfield (default: 0)
90 -n bool
92 if bool is non-zero, only recognise numbers (this is now obso‐
93 lete, use -C "0123456789")
94 The verbosity is specified as a bitfield:
96 1 print more info
98 2 list shapes of boxes (see -c) to stderr
100 4 list pattern of boxes (see -c) to stderr
102 8 print pattern after recognition for debugging
104 16 print debug information about recognition of lines to stderr
106 32 create outXX.png with boxes and lines marked on each general
108 OCR-step
109 The operation modes are:
111 2 use database to recognize characters which are not recognized
113 by other algorithms, (early development)
114 4 switching on layout analysis or zoning (development)
116 8 don't compare unrecognized characters to recognized one
118 16 don't try to divide overlapping characters to two or three
120 single characters
121 32 don't do context correction
123 64 character packing, before recognition starts, similar charac‐
125 ters are searched and only one of this characters will be
126 send to the recognition engine (development)
127 130 extend database, prompts user for unidentified characters and
129 extends the database with users answer (128+2, early develop‐
130 ment)
131 256 switch off the recognition engine (makes sense together with
133 -m 2)
134
138 Joerg Schulenburg (see http://jocr.sourceforge.net/ for EMAIL)
139 First version of man page by Tim Waugh <twaugh@redhat.com>
140
142 This man page documents gocr, version 0.41.
143
145 Report bugs to Joerg Schulenburg
146
148 More details can be found at /usr/share/doc/gocr-X.XX/gocr.html. Also
149 read /usr/share/doc/gocr-X.XX/README to learn, how to improve results.
150
152 gocr -v 33 text1.pbm
153 output verbose information, out30.png is created to see details
154 of recognition process
155 gocr -v 7 -c _YV text1.pbm
157 verbose output for unknown chars and chars Y and V
158 djpeg -pnm -gray text.jpg | gocr -
160 convert a jpeg file to pnm format and input via pipe
161Linux 29 Mar 2009 GOCR(1)