WO2002015118A1 - Method for automatically identifying a directed structure - Google Patents

Method for automatically identifying a directed structure Download PDF

Info

Publication number
WO2002015118A1
WO2002015118A1 PCT/DE2001/002993 DE0102993W WO0215118A1 WO 2002015118 A1 WO2002015118 A1 WO 2002015118A1 DE 0102993 W DE0102993 W DE 0102993W WO 0215118 A1 WO0215118 A1 WO 0215118A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
characterized
block
method according
structure
Prior art date
Application number
PCT/DE2001/002993
Other languages
German (de)
French (fr)
Inventor
Frank Müller
Original Assignee
Gavitec Ag
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to DE10040614A priority Critical patent/DE10040614A1/en
Priority to DE10040614.9 priority
Application filed by Gavitec Ag filed Critical Gavitec Ag
Publication of WO2002015118A1 publication Critical patent/WO2002015118A1/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/10544Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation by scanning of the records by radiation in the optical part of the electromagnetic spectrum
    • G06K7/10821Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation by scanning of the records by radiation in the optical part of the electromagnetic spectrum further details of bar or optical code scanning devices
    • G06K7/1093Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation by scanning of the records by radiation in the optical part of the electromagnetic spectrum further details of bar or optical code scanning devices sensing, after transfer of the image of the data-field to an intermediate store, e.g. storage with cathode ray tube
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/14Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/14Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
    • G06K7/1404Methods for optical code recognition
    • G06K7/1439Methods for optical code recognition including a method step for retrieval of the optical code
    • G06K7/1443Methods for optical code recognition including a method step for retrieval of the optical code locating of the code in an image
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/14Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
    • G06K7/1404Methods for optical code recognition
    • G06K7/1439Methods for optical code recognition including a method step for retrieval of the optical code
    • G06K7/1456Methods for optical code recognition including a method step for retrieval of the optical code determining the orientation of the optical code with respect to the reader and correcting therefore
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/40Analysis of texture
    • G06T7/41Analysis of texture based on statistical description of texture
    • G06T7/44Analysis of texture based on statistical description of texture using image operators, e.g. filters, edge density metrics or local histograms
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/70Determining position or orientation of objects or cameras

Abstract

In order to be able to automatically identify the location and orientation of a directed structure in a digital image, the invention provides a method in which the digital image is subdivided into preferably interconnected blocks. For each block, at least one texture characteristic is determined, which represents a measure for the structuring of the blocks along predetermined predominant directions, and parameters are determined from these texture characteristics. Said parameters indicate the location and the orientation of a directed structure in the digital image.

Description

A method for automatically detecting a targeted structure

The invention relates to a method for automatic detection of a location and an orientation of a directional structure in a digital image, in particular for determining the location and orientation of a bar code contained in the digital image.

Various methods for the detection and localization-related structures in images known. For example, there are methods in which from an image by applying an operator (z. B. a Laplacian-of-Gaussian operator, short LoG operator) is calculated, a second image whose pixel values ​​are not the brightness but the local contrast at the represent location of the respective pixels. In a second step of this second image is then examined to determine whether it contains along substantially straight and substantially parallel, lines of pixels having high values.

A disadvantage of such methods is the high computational effort, the elaborate processing units or long calls from values ​​times. Even the use of an operator may be too expensive for many applications. Even more, this applies to the subsequent evaluation of the second image. Other methods determine the two-dimensional autocorrelation function (ACF) or the two-dimensional Fourier spectrum within the local image areas. The values ​​of the ACF or the Fourier spectrum are then further processed to this information through the structuring of the respective areas to obtain. Even with these methods of computation complexity is very high.

The raised-looking structures in images can be barcodes. These barcodes are special optical codes, for example, of parallel bars (bars) are of varying thickness, by gaps (spaces) of different thicknesses are separated. The sequence of the thicknesses of bars and spaces represent stored in the bar code information is. Barcodes can be printed on product packaging and labels, or for direct marking use of various products. A key feature is the machine readable speed of barcodes. This means that it is possible with a special apparatus (barcode reader), which recover information stored in a bar code.

For reading bar codes are currently the two fundamentally different methods classes.

In one class, a time signal is formed by scanning the original with a laser beam, which is evaluated by a subsequent processing unit. This is called a so-called 1D method. In reading devices, which operate on the D-1 method, reading of the bar codes can only occur when the laser beam cuts the bar code perpendicular to the bars and spaces. When the relative position of the bar code reader and is not fixed from the outset, a reading of the code but should still allow (any position), it is necessary to scan the original along different directions several times (omnidirectional reading). According to this device is usually achieved with moving mirrors that deflect the laser beam so that the original is scanned hung in various places and along different RICH. The disadvantage here is the need to use moving parts, resulting in, among other limitations on the minimum size and the life.

In the other class, the scene is shot by a body having an electronic sensor and camera digitizes the sensor on the pro- jizierte image. The digital image is then transferred for evaluation to a decoding unit (2D) method. An advantage of 2D method is that one can do without moving parts in this bar code reader. A further advantage is present a digital image in the decoding unit which contains an image of the barcode.

The decoding can thereby randomly access the individual pixels of the image, making the class of possible Dekodiermethoden is substantially extended. In particular, the full repertoire of metal can methods of digital image processing in terms of a pre-processing of the image are used. It can cause errors or malfunctions of the code that may arise, for example, during the printing process, offset to a certain extent.

In the 2D code decoding process for one can distinguish whether a determination of the location and orientation of the barcode is carried out in the image (localizing procedures) or not.

Locating methods are particularly advantageous when the position and / or orientation of the bar code within the field of view of the camera can not be accurately predicted and the bar code to be read also (omnidirectional reading).

For non-localized process, the ability to omnidirec--directional reading by evaluation of the image along a plurality of virtual scan lines are obtained. It is therefore gebil- det from the image LD signal by the gray values ​​of the image at different locations along different directions are read out. The thus formed LD signal is then evaluated with basically similar procedure as a generated by a laser scanner time signal. is disadvantageous here is that the gray values ​​of the image are counted along a plurality of lines off and thus the one hand multiplies the calculation effort on the other hand the risk of incorrect readings is increased. In a known method which avoids these drawbacks by a location and orientation of the bar code, additional markers are mounted in the immediate vicinity of the bar code. These marks are detected in a first step by a correspondingly modi- ed reader. Since the markers are at a predetermined distance to the barcode, and the barcode is located by locating the markers. In this method, it is disadvantageous that it requires printed markings. Since these markings do not form part of a generally applicable standards, they are not used consistently.

Moreover, the additional markings will also need an additional printed surface. Especially with the marking of goods (EAN / UPC), the printable area is valuable as an advertising medium, which is why the use of additional markers previously could not prevail.

Because bar codes are composed of parallel lines, they represent in the sense of digital image processing directed illustrate structures and it is possible in principle, originating from the digital image processing method for localization-related structures (texture analysis) to also use for locating barcodes. These methods require JE but the use of special operators (z. B. LoG operator), the determination of two-dimensional correlation functions or the determination of the two-dimensional Fourier spectrum, so the application of computationally expensive algorithms and are therefore not economically suitable for locating barcodes.

It is therefore an object of the present invention to provide a method which is suitable for automatically detecting a location and a O- rientierung a directional structure in a digital image, in particular for recognition of a barcode included in a digital image, and for at least approximate determination of the location and can economically inserting the orientation of this bar code.

According to the invention the above object is achieved in that in a method for the automatic detection of a location and an orientation of a directional structure in a digital image, in particular for determining the location and orientation of a bar code included in the digital image, the digital image in preferably contiguous blocks is subdivided, and at least one texture is characteristic determined for each block, which is a measure of the structure of the blocks along predetermined preferred directions, and parameters are determined from the texture features that indicate the location and orientation of a directional structure in the digital image ,

Advantageously, the picture here is a delesegerät captured by a camera-based co-image of a real scene containing a barcode among others. Under an electronically present or digital picture of a matrix of picture elements (pixels) to-is understood sammengesetztes object, wherein each pixel (a place (usually in the form of a row index and the column index), and the brightness at the location of the pixel value indicative value of the pixel) is associated. However, it is also produced in other ways, be automatically analyzed in electronic form present images with the presented method.

Particularly advantageous in the inventive method is that the computational complexity compared to known processes directed to the localization structures, many times reduced. Application of this method for locating barcode is economically very interesting by this reduced computational effort.

A further advantage, the process of the invention brings with it is, that no additional markings are required with respect to the barcode having surface which were needed in conventional methods for localization and orientation determination for the barcode. Therefore, no additional space for the marks is required, making this area stands as a potential advertising space. This is for example particularly interesting if the packaging of a product is relatively My.

Advantageous in the process is also that only a single line needs to be evaluated even after localization. In addition, the concentration reduces the risk of incorrect readings on those image regions containing a high probability of barcodes.

To advantageously locate a directional structure, the digitalized le image is divided into areas, the so-called blocks. Here, the blocks are arranged preferably contiguous. To simplify the process further, it is advantageous if the blocks have a rectangular shape, preferably a square shape.

An embodiment variant of the method provides that for each of these blocks is preferably a predetermined number of first direction-sensitive characteristics, calculates the so-called "first-order features." This "first-order features" are hereinafter referred to as "first texture features".

The calculation rule for this "first order features" is described by so-called characteristic features. A characteristic feature is an assignment rule which assigns a block in an unambiguous way a number. All characteristic features are designed so that a block will be assigned the higher the values, the greater the gray value differences of the block along a predetermined preferred direction. There is thus defined a set of feature functions, each of the characteristic functions of a preferential direction can be assigned from this set, along which an image block is evaluated the feature functions are preferably designed. that the associated preferred directions cover the totality of all directions evenly as possible.

In particular, the characteristic function can be selected so that two loading arbitrarily selected different feature functions also have different preferred directions. Here, a "first-order feature" by using a feature function is determined on a block. So you will receive for each block and for each feature function value.

It is advantageous if the ratio of the number of pixels per block to the number of characteristic features is significantly greater than 1 because the number of features per image then is significantly lower than the number of pixels of the image, thus increasing the complexity of the following calculation considerably reduced.

In order to gain additional savings in computation time and memory access time, it is advantageous to evaluate each case only a part of the pixels of a block.

One can imagine doing the feature function as a "gauge" that measures the "activity" of an image or Bildausschnit- tes along the preferred direction. The texture feature is then the "rash" of the direction-sensitive measuring device "feature" function. Includes a block a directional structure, a textile is turmerkmal which extends substantially parallel to the structure, have a relatively low value. One feature that is substantially perpendicular to the structure is, however, have a relatively high value. A directional structure within a block is thus detected in two different ways.

First, a directional structure expressed by high values ​​of a characteristic function whose preferred direction is substantially perpendicular to the structure. Secondly, a directional structure expressed by low values ​​of a characteristic function whose preferred direction extends parallel to the GR sentlichen structure.

These already set at least two ways to realize a directional structure in one block. On the one hand, one can use very low values ​​of a texture feature as an indication of the presence of a targeted structure along the associated with the texture feature of the preferred direction, on the other hand, one may use particularly high values ​​of a texture feature as an indication of the presence of a targeted structure perpendicular to the preferred direction.

In order to eliminate the influence of non-directional structures, a new can with the inventive method from two "first-order features", whose characteristic features are characterized by substantially mutually perpendicular preferred directions are determined "second texture feature". The linkage of any two "first order features" is then preferably designed such that the resulting "second texture feature" assumes high values, when the value of a "feature of first order" high and simultaneously the value of the other "feature of first order" is low , A thus by offsetting two "first-order features" certain "second texture feature" called "second-order characteristic" hereinafter also.

A "second-order characteristic" thus takes a high value, for example, when the block along the preferred direction has a high activity and simultaneously perpendicular to the preferred direction has a low activity. It is achieved in this way, among other things, that neither featureless areas still undirected structured regions to a high value of a "second-order feature," and thus to a misinterpretation (supposed detection of a nonexistent oriented structure) may result.

A "feature of the second order" is understood to be a characteristic whose value is a measure of the probability of Vorlie- gens a directional structure perpendicular to a preferred direction, regardless of whether it only by offsetting two "first order features" or in a step was calculated.

Belonging respectively to the same direction "second texture features" are entered for example in so-called feature maps, preferably in each case at the point at which the corresponding block of the input image is located. The feature card is in this case for example a reduced image, the pixels of which preferably not more the indicating brightness of the original image, but the probability of the existence of a directional structure of a particular direction.

, It is advantageous that preferably not each pixel of the original image is assigned to value a separate feature, but include the various characteristic values ​​to the blocks and thus the feature map, for example, correspondingly smaller than the original image.

It is also advantageous that the evaluation preferably based only on predetermined features but not to the image gray values. Thus, a significantly smaller amount of computation can be achieved when, for example, an image produced with a LoG operator image. This is partly due to the data reduction (the number of features is smaller than the number of pixels in the input image), on the other hand, however, is this is also the fact that the characteristic functions are constructed so that they allow conclusions on the local orientation of the individual image areas.

These advantages evaluation results can be more efficient and thus achieve more economic. As a result of this process are, among other parameters that specify a statement about the location and orientation or an angular position of a directional structure in a digital image. Furthermore, a probability for the existence of a directional structure can be determined by the parameters.

One embodiment provides to put first those "second order characteristics" in preferably each feature map to zero, the characteristic values ​​are, for example, below a suitably chosen threshold, and to use for example, only those features for further evaluation, the value of which is really positive. The height the threshold is substantially influence friendliness to the Detektionsempfind-.

In this method, it is advantageous to select a low threshold, if one wants to detect, for example, also oriented structures with low contrast. Conversely, it is advantageous to select a high threshold, when the structure to be recognized, for example, a high contrast has (eg. As in a properly printed barcode), but is expected at the same time with high contrast textured interference in the environment of the desired directional structure.

In this case is preferably first selected from all feature maps a feature map, which indicates a high probability of the existence of egg ner directional structure in the image. It is also advantageous to select those feature maps, the values ​​of which for example have the biggest mean value. The thus selected "best" feature map is further examined to confirm the existence of a directed structure or be discarded. The location of a present in the image-looking structure, by the determination of a location parameter (z. B. average, weighted average, Mediän) of advertising the feature map estimated. the angular position of the directional structure results roughly from the already associated with the selected feature map preferential direction. a more accurate estimate of the orientation of the structure is preferably determined by the application of a regression method from the feature map.

Advantageous for determining the first-order feature map, among others, is a block-wise analysis of gray value differences of pixel pairs along the selected preferred direction, because this method can be implemented, for example, to the formation of two-dimensional autocorrelation function or the two-dimensional Fourier spectrum with significantly less computational effort.

To minimize the computational as well as the time spent on, it is advantageous to select those initially preferred directions in the evaluation of texture features, their associated texture features Las best explained by the existence of targeted structure in the image sen. By this method, the data can continue to reduce expenses. Further advantages, objects and features of the present invention are illustrated by reference to the following explanation of the attached drawing, in which by way of example, the recognition method is shown.

It shows,

the figure is a schematic representation of the detection method based on an embodiment.

In the embodiment which is shown in the figure, an image pickup unit is connected to a computing unit. The Bildaufnah- meeinheit will deliver a digital image of size 640 x 480 image points, typically with 256 shades of gray to the computing unit.

The image is then divided into first 32 x 20 x 15 non-overlapping blocks of size 32 pixels. Next four "first-order features" are determined for each of these blocks. The calculation formula is described every "first-order feature" by a respective feature function.

Each feature function has in this case each image block in an unambiguous manner to a number, and is defined by two parameters, namely by

1 a so-called clique vector (delta_x, delta_y)

2. an increment DSAMPLE. The components of the vector clique are integers (positive or negative), and the increment DSAMPLE is a natural number.

For determining a first-order feature all pixel pairs ((25 x y 2)) are formed, which satisfy the following conditions simultaneously:

1. they are entirely within the block under consideration,

2. it is X j - x 2 = Delta_x and y x - y 2 = delta_y,

3. x x and y are divisible by DSAMPLE.

The Condition 2 states that the pixel pairs thereby drawing off, on the one hand, the connecting line between the pixels of the pair along the given by the clique vector preferred direction of running and that, secondly, the distance of any two points of a point pair is constant.

Condition 3 makes it possible to select only one of the possible according to Condition 2 within a block point pairs, thus enabling a reduction of the computational effort for determining the respective texture feature.

The next step is to determine the "first order characteristic". In this case, it is determined for each pixel pair, which satisfies the above conditions of the waste solutbetrag the difference or the square of the difference between the gray values ​​of the two points. Then the values ​​thus obtained are summed ,

During implementation, four clique vectors are used, namely

c = (3,0) (which is the horizontal preferential direction), c 2 = (0.3) (this is the vertical preferential direction), c 3 = (2,2) (which is the diagonal preferred direction), c 4 = (-2.2) (this is the preferred anti diagonal direction).

Here DSAMPLE is set to 4, thereby resulting in a savings in computing time of approximately 16 yields over a DSAMPLE value of Figure 1.

In a next step, the "first order features" in so-called feature maps are entered.

In the implementation example of the thus obtained are registered "first order features" that is, in four (number clique vectors) feature maps which have 15 (number of blocks = number of features per clique vector) each the size of 20 x.

The following are the designated "feature maps first order" with T_n (x, y), where n is the values ​​1 to 4, x is from 1 to 20 and y assume the values ​​1 to 15 °. Thus, for example, designated T_2 (3,4) the

"First-order feature" for the third block from the left in the fourth row for the clique vector c. 2 The feature T_2 (3,4) are in this example, the activity of the block at the position (3,4) along a vertical direction at.

In a further step, the determined "feature maps of the second order" from the "feature maps first order".

To determine a "feature map second order" M_i a is in each case offset "feature map of the first order" T_i with another "feature map of the first order" TJ The allocation is made therein with a feature map TJ whose clique vector C j perpendicular to the clique vector C j of ". feature map of the first order is. "

In the example, each c are: and c 2 and c 3 and c perpendicular to each other. 4 The type of calculation is determined by two parameters and THRESHOLD FACTOR and blockwise as follows ben described. First, be determined as follows, the "second-order feature maps":

M_L (x, y) = T_l (x, y) - FACTOR * T_2 (x, y)

M_2 (x, y) = T_2 (x, y) - FACTOR * T_l (x, y), M_3 (x, y) = T_3 (x, y) - FACTOR * T_4 (x, y),

M_4 (x, y) = T_4 (x, y) - FACTOR * T_3 (x, y). In a next step, the threshold value is determined, which depends on the particular application. Here, the threshold value is substantially influenced by the selected block size and the gray level resolution of the original image. In this step, all entries are set in all feature maps to zero if they are smaller than this threshold value precisely.

Typical values ​​for the factor are in the range 1.5 ... 2.5. After that, the feature maps are sorted. This is done so that for each feature map, the number of non-zero entries is overall one and the feature cards are sorted by this number in descending order. Thus, the first of the feature maps obtained in the list, most of zero entries and is therefore considered as the first candidate.

Thereafter, the focus of the feature maps to be viewed is determined (the first moment in the x and y-direction). The center of gravity (x s, y s) thereby has real-valued coordinates in the range x s = 1 ... 20, y s = 1 ... 15, and is on integer values round (x s, y s) rounded.

If the considered feature map round at the point (x s, y s) contains a non-zero value, continuing with this feature map as the current map.

If the considered feature map, however, contains a zero instead of round (x s, y s), the focus of the next characteristic map is determined by the list. It is then checked whether the card contains a non-zero value at the point of gravity.

In this way, all the candidates of the list are processed until either a card was found that at the point of gravity has a non-zero value, or until the list has been completed without success.

In the latter case, the process is terminated because there is no directional structure was found.

Otherwise, from the center of gravity (x s, y s) of the current Merkmalskar- te determined by scaling with the block size of the center of the bar code (based on the size of the original image). For the characteristic map is determined current best-fit line by linear regression one through the center of gravity (y s x s).

This regression line is substantially perpendicular through the directional structure. When the directional structure is now a bar code, which is to be decoded, it can be used straight as a virtual scan line, as it fully intersects with high Walirscheinlichkeit the barcode.

The method described herein for the automatic recognition of an or- tes and an orientation of a directional structure in a digital

Picture is characterized by reduction of the data contained in the requested original image data, so that a considerably lower computational effort must be operated in order to locate a directional structure. It is advantageous that the digital image is divided into adjacent blocks, and these blocks are written by loading feature functions so that not each pixel enters in the input image in the calculation, but only the number of texture features.

A further advantage is that the possibility of erroneous readings is reduced by the fact that the method, for example, makes it possible to focus on those areas of the image ness with high probability comprise a bar code.

Claims

claims:
1. A method for automatic detection of a location and an orientation of a directional structure in a digital image, in particular for determining the location and orientation of a bar code included in the digital image, characterized in that the digital image is divided in preferably contiguous blocks, and at least one texture feature is determined for each block, which is a measure of the structure of the blocks along predetermined preferred directions, and parameters are determined from the texture features that the location and
indicate orientation of a directional structure in the digital image.
2. Detection method according to claim 1, characterized in that the blocks have a rectangular, preferably square form.
3. Detection method according to one of claims 1 or 2, characterized in that, for each image block a plurality of "first texture features" determines that take a so higher value, the larger the sum of the absolute values ​​of the gray-scale difference precipitates of pixel pairs, the pixel pairs for each "first texture feature" are formed from image points which lie within the considered block whose Verbindungsli- never lie along the the considered "first texture characteristic" associated preferred direction, and the distance a fixed value or a value of a fixed amount of corresponding values.
4. Detection method according to claim 3, characterized in that only a proportion of the pixels within the block under consideration, are the connecting line along the associated preferred direction and the pixel pitch corresponding to a predetermined value or a value from a defined set of values, are evaluated.
5. Detection method according to one of claims 1 to 4, characterized in that for determining the value of "a second texture characteristic" of a block, the values ​​of two "first texture features" of the same block are used along a measure of the structure of the block specify two substantially mutually perpendicular preferred directions.
6. Detection method according to one of claims 1 to 5, characterized in that the calculated "first texture features" are evaluated.
7. recognition method according to any one of claims 1 to 6, characterized in that in the evaluation of texture features initially that the preferential direction is chosen whose zugehöri- ge texture features can best be seen with the existence of a directional structure in the image, and in which then only those texture features that belong to the selected preferred direction, are used to determine the parameters that indicate the probability of the existence of a directional structure and / or the location of a targeted structure and / or the position of a directional structure in the digital image.
8. Detection method according to one of claims 1 to 7, characterized in that the image is divided into overlapping regions.
9. recognition method according to any one of claims 1 to 8, characterized in that for adjusting the detection sensitivity, a threshold value in the evaluation of texture features is determined.
10. recognizing method according to any one of claims 1 to 9, characterized in that the parameters specify a probability for the existence of a directional structure in the digital image.
PCT/DE2001/002993 2000-08-16 2001-08-13 Method for automatically identifying a directed structure WO2002015118A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
DE10040614A DE10040614A1 (en) 2000-08-16 2000-08-16 A method for automatically detecting a targeted structure
DE10040614.9 2000-08-16

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
DE10193420A DE10193420D2 (en) 2000-08-16 2001-08-13 A method for automatically detecting a targeted structure
AU8379201A AU8379201A (en) 2000-08-16 2001-08-13 Method for automatically identifying a directed structure
EP01962610A EP1312034A1 (en) 2000-08-16 2001-08-13 Method for automatically identifying a directed structure

Publications (1)

Publication Number Publication Date
WO2002015118A1 true WO2002015118A1 (en) 2002-02-21

Family

ID=7653010

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/DE2001/002993 WO2002015118A1 (en) 2000-08-16 2001-08-13 Method for automatically identifying a directed structure

Country Status (4)

Country Link
EP (1) EP1312034A1 (en)
AU (1) AU8379201A (en)
DE (2) DE10040614A1 (en)
WO (1) WO2002015118A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7728959B2 (en) 2003-06-21 2010-06-01 Aprilis, Inc. Acquisition of high resolution biometric images
US7734729B2 (en) 2003-12-31 2010-06-08 Amazon Technologies, Inc. System and method for obtaining information relating to an item of commerce using a portable imaging device
US8538095B2 (en) 2003-06-21 2013-09-17 Aprilis, Inc. Method and apparatus for processing biometric images
US8631089B1 (en) 2010-12-14 2014-01-14 Brilliance Publishing, Inc. Previewing audio data associated with an item

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE10351861A1 (en) 2003-11-06 2005-06-09 Deere & Company, Moline Method and steering system for automatically steering an agricultural machine

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5373147A (en) * 1992-09-16 1994-12-13 International Business Machines Corporation Apparatus and method for detecting line segment direction
EP0917080A1 (en) * 1997-11-17 1999-05-19 Datalogic S.P.A. Method of locating highly variable brightness or colour regions in an image
EP0984390A2 (en) * 1998-08-31 2000-03-08 Xerox Corporation Apparatus for providing location information in a multidimensional address space
EP0996079A1 (en) * 1998-10-23 2000-04-26 Datalogic S.P.A. Method for locating codes in bidimensional images

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5329105A (en) * 1992-08-10 1994-07-12 United Parcel Service Of America, Inc. Method and apparatus for determining the width of elements of bar code symbols
US5969325A (en) * 1996-06-03 1999-10-19 Accu-Sort Systems, Inc. High speed image acquisition system and method of processing and decoding barcode symbol
US5974200A (en) * 1997-03-20 1999-10-26 Matsushita Electric Industrial Company, Ltd. Method of locating a machine readable two dimensional barcode within an image
US6082619A (en) * 1998-12-16 2000-07-04 Matsushita Electric Industrial Co., Ltd. Method for locating and reading a two-dimensional barcode

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5373147A (en) * 1992-09-16 1994-12-13 International Business Machines Corporation Apparatus and method for detecting line segment direction
EP0917080A1 (en) * 1997-11-17 1999-05-19 Datalogic S.P.A. Method of locating highly variable brightness or colour regions in an image
EP0984390A2 (en) * 1998-08-31 2000-03-08 Xerox Corporation Apparatus for providing location information in a multidimensional address space
EP0996079A1 (en) * 1998-10-23 2000-04-26 Datalogic S.P.A. Method for locating codes in bidimensional images

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7728959B2 (en) 2003-06-21 2010-06-01 Aprilis, Inc. Acquisition of high resolution biometric images
US8538095B2 (en) 2003-06-21 2013-09-17 Aprilis, Inc. Method and apparatus for processing biometric images
US7734729B2 (en) 2003-12-31 2010-06-08 Amazon Technologies, Inc. System and method for obtaining information relating to an item of commerce using a portable imaging device
US8631089B1 (en) 2010-12-14 2014-01-14 Brilliance Publishing, Inc. Previewing audio data associated with an item
US9256395B1 (en) 2010-12-14 2016-02-09 Brilliance Publishing, Inc. Previewing audio data associated with an item

Also Published As

Publication number Publication date
EP1312034A1 (en) 2003-05-21
DE10193420D2 (en) 2003-07-03
AU8379201A (en) 2002-02-25
DE10040614A1 (en) 2002-02-28

Similar Documents

Publication Publication Date Title
US5446271A (en) Omnidirectional scanning method and apparatus
US6754369B1 (en) License plate reading apparatus and method
US7181066B1 (en) Method for locating bar codes and symbols in an image
CA2217458C (en) Method for locating the position and orientation of a fiduciary mark
US6053407A (en) Maxicode data extraction using spatial domain features
US20050175221A1 (en) Method of determining usability of a document image and an apparatus therefor
US5245674A (en) Image processing using distance as a function of direction
EP2168079B1 (en) Method and system for universal lane boundary detection
US7747039B2 (en) Apparatus and method for automatically detecting objects
DK1738136T3 (en) Measuring apparatus and method in a distribution system
EP0629970B1 (en) Method for detecting monetary amounts in binary images
US7437243B2 (en) Detecting device and method to detect an object based on a road boundary
US20020118874A1 (en) Apparatus and method for taking dimensions of 3D object
US5262623A (en) Method and apparatus for distinguishing a preferred bar code or the like
CA2196875C (en) Method and apparatus for determining the fine angular orientation of bar code symbols in two-dimensional ccd images
EP0569962B1 (en) Method and apparatus for processing a two dimensional digital image
EP0894311B1 (en) Omnidirectional barcode locator
Moghadam et al. Fast vanishing-point detection in unstructured environments
EP0685809A2 (en) Method and apparatus for decoding two-dimensional bar code
US5504319A (en) Method and system for bar code acquisition
EP0582548B2 (en) Method and apparatus for determining mis-registration
Hull Document image skew detection: Survey and annotated bibliography
US7430303B2 (en) Target detection method and system
US20030035098A1 (en) Pose estimation method and apparatus
US5120940A (en) Detection of barcodes in binary images with arbitrary orientation

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NO NZ PL PT RO RU SD SE SG SI SK SL TJ TM TR TT TZ UA UG US UZ VN YU ZA ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): GH GM KE LS MW MZ SD SL SZ TZ UG ZW AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
WWE Wipo information: entry into national phase

Ref document number: 2001962610

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 2001962610

Country of ref document: EP

REF Corresponds to

Ref document number: 10193420

Country of ref document: DE

Date of ref document: 20030703

Kind code of ref document: P

WWE Wipo information: entry into national phase

Ref document number: 10193420

Country of ref document: DE

NENP Non-entry into the national phase in:

Ref country code: JP

WWW Wipo information: withdrawn in national office

Ref document number: 2001962610

Country of ref document: EP