Full text: Papers accepted on the basis of peer-review full manuscripts (Part A)

ISPRS Commission III, Vol.34, Part 3A „Photogrammetric Computer Vision‘, Graz, 2002 
  
  
THE RESOLUTION ENHANCEMENT BY 
SOFTWARE METHOD OF RANDOM SHIFT 
Tomas Ponik, Jaroslav Placek 
FMFI UK, Bratislava 
Abstract 
The goal of the paper is to describe one 
method of software resolution enhancement of records 
obtained by CCD cameras or other devices able to 
view and digitalize the image. The paper is divided 
into two parts. The first one explains the theoretical 
principle of the random shift method. The second part 
is dedicated to application HimRes which tests these 
method and creates resulti images with enhanced 
resolution. 
Keywords: CCD camera, linear properties of the 
CCD camera, light-sensitive element, raster, center of 
gravity 
Used terms: pixel, object, record, random shift, 
detection object, resulting image, actual pixel, 
covering, overlap (definitions are explained bellow). 
Introduction 
The CCD camera is nowadays a well-known 
devide used in many areas of science, development 
and industry. It is cheap and easy to use. Analog CCD 
cameras produce analog signal (PAL or NTSC) as the 
output which has the resolution of 768x576 pixels. 
This can be enough for simple image analysis but in 
many cases user wants to see further, to see details and 
other features in the image which is impossible with 
the use of analog cameras. The digital CCD camera 
with resolution up to 1280x1024 exists already but it's 
price is 40 times higher than ordinary analog one. 
There is a way to enhance the resolution of 
camera image by snapping more pictures of the same 
scene and computing the new image using suitable 
features of these image series. 
The explanation of used terms: 
CCD camera — The device for viewing of the image. 
CCD-elements — The elements of the CCD camera 
sensitive to light intensity which are created by MOS 
condensors or PN transitions. 
Light-sensitive element (shortly element) — CCD- 
element, which represents pixel in raster image. 
Pixel — Two-dimensional image element, which has a 
color. The definition of color is bellow. 
Scene — The part of the real world viewed by te CCD 
camera or other device able to record the image. 
Object — The part of the scene. 
Record - Scene viewed by the CCD camera and 
digitalized to 2D space. Every record is with respect to 
other obtained records shifted, resp. rotated in the 
image plane. 
Random shift — Informations of record positions in 
choosed coordinate system are not contained in the 
input data. The record capture is performed by random 
shift of the CCD camera, resp. of the viewed object. 
Detection object — The circle or other geometric solid 
invariant to the operation of rotation which is used by 
determination the relative position of the record with 
respect to other records. 
Image with compressed sampling (shortly resulting 
image) — Raster image with higher pixel density as on 
the records. 
Actual pixel in resulting image (shortly actual pixel) — 
Actually processed pixel in resulting image. 
The pixel color — Integer value representing the 
illuminance recorded by the image element. During 
creation of resulting image the domain for these 
values is the set of real numbers between 0 and 255. 
The cover of actual pixel by the record pixels parts — 
Parts of the record pixels cover actual pixel if and only 
if the actual pixel surface is created by the unification 
of these parts (Figure 1). 
Overlap of actual pixel by the record pixels — Record 
pixels overlap the actual pixel if and only if their parts 
cover actual pixel (Figure 1). 
  
pixel 
  
Figure 1 Cover and overlap of actual pixel 
The coordinate system of image plane (shortly 
Choosed coordinate system) —The two-dimensional 
cartesian coordinate system is used in this paper. It is 
oriented according to the first captured record (Figure 
2). The coordinate axis are parallel to record rows and 
columns. The center of the coordinate system is placed 
in the intersection of first row and first column of 
record. 
For easier explanation only grayscale images 
are considered with black color evaluated as 255 and 
white color as 0. In the next explanation the resolution 
enhancement of factor two is considered. 
A - 258
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.