libcvautomation/docs/man/man1/cva-match.1
2012-08-03 15:04:34 -04:00

61 lines
1.9 KiB
Groff

.TH "cva-match" 1 "3 Aug 2012" "Version 1.4" "libcvautomation" \" -*- nroff -*-
.ad l
.nh
.SH NAME
cva-match \- \fBAuthor:\fP
.RS 4
Bradlee Speice <bspeice@uncc.edu>
.RE
.PP
.SH "Usage:"
.PP
This program uses OpenCV in order to recognize an image within an image. The return code is 0 for at least one successful match, and 1 otherwise.
.SH "Example Usage:"
.PP
Match two images against the root X11 window:
.PP
cva-match --x-root -s '<image_name>' -s '<image_name_2>'
.PP
Match a root and sub image with a custom tolerance and search method
.PP
cva-match -r '<root_image>' -s '<image_name>' -o 75 -m 1
.SH "Full Options:"
.PP
-h, --help: Display this usage message.
.PP
-u, --usage: Display this usage message.
.PP
-r, --root-image: Location of the root image to compare against.
.PP
-s, --sub-image: Location of the sub-image to find in root.
.PP
-p, --separator: Separator of the X and Y coordinates.
.PP
-t, --tolerance: Set how strict the match is - 100 is recommended lowest value.
.PP
\fBNote:\fP
.RS 4
When using CCORR or CCOEFF tolerance works in opposite direction, so -50 is recommended highest value.
.RE
.PP
-m, --search-method: Set which method is used to search for sub-images.
.PP
\fBNote:\fP
.RS 4
Search Methods: CV_TM_SQDIFF = 0, CV_TM_SQDIFF_NORMED = 1, CV_TM_CCORR = 2, CV_TM_CCORR_NORMED = 3, CV_TM_CCOEFF = 4, CV_TM_COEFF_NORMED = 5
.RE
.PP
-x, --x-root[=DISPLAY]: Set the root image to come from X11
.PP
\fBNote:\fP
.RS 4
The DISPLAY variable is optional, not specifying it will cause X to use the default display (not specifically :0.0)
.RE
.PP
-c, --center: The output points should be centered on the sub-image, rather than the top-left corner.
.PP
-o, --sane-tolerance: Set the tolerance using a scale of 1-100, rather than INT_MIN to INT_MAX (100 ~= INT_MAX)
.SH "Contact Information:"
.PP
Questions? Comments? Concerns? Suggestions? Send all feedback to Bradlee Speice at <bspeice@uncc.edu>