Scan-assisted mobile telephone

Abstract

The invention relates to a method for scanning, in particular documents with the aid of a camera (K). The inventive method consists in combining an image sequence or a sequence of partial images in order to obtain a final image, projecting a reference model (F) onto the camera vision field and in compensating representational differences of different images on the basis of geometrical information of the model (F).

Description

The invention relates to a method for scanning, in particular documents as well as to a mobile telephone according to the preamble of claim 2.

The use of a camera, typically with a low resolution, for scanning documents is known. This involves the camera and/or the mobile telephone being moved over the document at a short distance from it. A final image is generated from the images and/or partial images recorded in this way.

The images are recorded for example using a high scanning frequency of typically 15 or 30 Hertz. For this reason, adjacent images have an extremely high image overlap. A high resolution final image can be continuously established by adjoining narrow film strips from the respective image centre and allowing for the camera displacement.

Small variations in the distance between the camera and the document and a slight tilting of the camera which occurs with a hand-held camera nevertheless result in extremely troublesome changes in size in the document content displayed. The size changes must be adjusted to each other in a first processing step, so that the high resolution final image can be constructed without errors. The determination of the position of the camera sensor with respect to the document, particularly in terms of the distance and tilting, is a fundamental and important condition for fault compensation. The calculation of the relative position information from successive images is a known approach, which is extremely complex and provides results which are mostly very unstable. In general, tiltings of the camera in relation to the document can only be determined with great difficulty, on the basis of the recorded image information by means of image processing.

The object of the invention is to facilitate the scanning of documents using a scanning method and a mobile telephone.

This object is achieved according to the invention for a method by means of the features specified in claim 1 and for a mobile telephone by means of the features specified in claim 2.

The method and/or mobile telephone according to the invention allows for the rapid and robust determination of the relative position of the camera sensor in respect of the document.

In accordance with the invention, a reference model is projected onto the visual field of the camera. The evaluation of the geometry of the pattern in the images enables the compensation of the representational differences in the images.

A small reference model is permanently projected onto the same position within the visual field of the camera and is evaluated in the recorded image by means of an image processing algorithm. The position of the sensor in relation to the document is uniquely determined in a three-dimensional space from the geometric information of the reference model projected onto the document surface and the reference model represented in the sensor. The use of a reference model permits the extraction of the necessary geometric information for compensating the representational differences in images to be very robust and rapid using the methods of digital image processing and to be effected independently of the content of the scanned document.

A suitable reference model is for example a small reticule, which is permanently projected onto the same position within the visual field of the camera. The position of the sensor in relation to the document is distinctly determined in a three dimensional space from the information of the reticule parameter (distance) and the relative length of the reticule axis (tipping in x and y direction).

SUMMARY OF THE ADVANTAGES OF THE METHOD

• • Rapidity and robustness in the determination of the relative position of the sensor to the document
  • Extremely precise determination of the distance and the tilting of the sensor irrespective of the content of the scanned document

The invention is described below in more detail with reference to an exemplary embodiment shown in the drawing.

The method for the position determination of a camera K in relation to a document (not illustrated in more detail), as shown in FIG. 1, is realized for example on a mobile telephone or a mobile radio device which has the following features:

• • Camera K for example on the rear side of the mobile telephone MT for recording images from the surroundings
  • Processor for processing image information and for determining the relative sensor position
  • Light-emitting diode L on the rear side of the mobile telephone MT for projecting a reticule F into the visual field of the camera (K).

The mobile telephone MT moves across the document at a slight distance therefrom in order to scan a document for example. The characteristic features of the projection of a reference model F for determining the relative sensor position are roughly sketched on an exemplary design.

Case a) in comparison with case b)

In cases a) and b) the camera is parallel to the document surface. The distance of the camera to the document is larger in case a) than in case b). The projected reticule F in the camera image is thus smaller in case b) than in case a). The reticule size is directly linked to the sensor distance and can be calculated in each instance from the detail screen of the camera.

Case c)

The camera K and/or the camera sensor is tilted in respect of the master document. The position of the sensor in a 3-dimensional space in relation to the document can be distinctly determined from the absolute size of the reticule in conjunction with the information about the relative length of the reticule axis.

Claims

1. Method for scanning, in particular documents, using a camera (K), an image sequence or a sequence of partial images being combined into a final image, in which a reference model (F) is projected onto the visual field of the camera, and in which representational differences of different images are compensated on the basis of the geometrical information of the model (F).

2. Mobile telephone (MT) with a camera (K) and with an image processing device for generating a final image from the images or partial images obtained during a scanning process, characterized by a projection device (L), by means of which a reference model (F) can be projected onto the visual field of the camera, and means of compensating for representational differences of the different images on the basis of the geometrical information of the model (F).