JP2011186777A - Bar-code reading device and bar-code reading method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a bar-code reading device capable of decoding a bar code with high accuracy, from a bar-code image of low resolution, and to provide a bar-code reading method. <P>SOLUTION: The bar code reader includes a scanning data generating unit for generating a plurality of pieces of scanning data of the bar code; a high resolution image processing unit for generating one data sequence of resolution higher than the respective resolutions of the plurality of pieces of scanning data, on the basis of the plurality of pieces of scanning data; and a decoding unit for decoding that decodes the information of the bar code from the data sequence generated. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a barcode reading apparatus and method for reading a barcode from a form or an object, and more particularly to a barcode reading apparatus and a barcode reading method for reading a barcode from a low resolution image.

With the recent development of IT, the demand for barcodes with a large amount of information is increasing, and the density of barcodes is increasing. Conventionally, when a high-density barcode is read by a barcode reader, a scanner with a high resolution (high resolution) corresponding to the fineness (module width, etc.) of the barcode to be read has been required. For example, in order to read a barcode of the EAN128 specification in which the minimum module width is defined as 0.167 mm, a scanner having a resolution (resolution of 300 dpi or more) at least about twice that is required.
A technique for restoring high frequency components of an original signal having a Nyquist frequency or higher from a plurality of sets of digital data having different sampling positions to obtain high resolution data with little blur has been disclosed (see Patent Document 1).

JP-A-8-336046

Here, a high-resolution scanner is expensive, and there are problems such as that it takes time to scan to obtain a high-resolution image, and the data capacity of the obtained image increases.
In view of the above, an object of the present invention is to provide a barcode reading apparatus and a barcode reading method capable of decoding a barcode with high accuracy from a low-resolution barcode image.

  A barcode reader according to an aspect of the present invention includes a scan data generation unit that generates a plurality of scan data of a barcode, and a resolution higher than the resolution of each of the plurality of scan data based on the plurality of scan data. A high-resolution processing unit that generates one data series, and a decoding unit that decodes the barcode information from the generated data series.

  A barcode reading method according to an aspect of the present invention includes a step of generating a plurality of barcode scan data, and one data having a resolution higher than the resolution of each of the plurality of scan data based on the plurality of scan data. Generating a sequence; and decoding the barcode information from the generated data sequence.

  According to the present invention, it is possible to provide a barcode reading apparatus and a barcode reading method capable of decoding a barcode with high accuracy from a low resolution barcode image.

1 is a block diagram illustrating a barcode reading apparatus 10 according to a first embodiment of the present invention. It is a figure showing the image data of the form acquired by the photoelectric conversion part 11. FIG. 6 is a schematic diagram showing a result of scanning by a barcode scanning unit 14. FIG. It is a block diagram which shows the barcode reader 20 which concerns on the 2nd Embodiment of this invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
(First embodiment)
FIG. 1 is a block diagram showing a barcode reading apparatus 10 according to the first embodiment of the present invention. The barcode reader 10 is a device that reads a barcode printed on a form.

  The barcode reader 10 includes a photoelectric conversion unit 11, an image storage unit 12, a barcode detection unit 13, a barcode scanning unit 14, an image reconstruction unit 15, a decoding unit 16, and an output unit 17. Hereinafter, the photoelectric conversion unit 11 and the like will be described in order.

  In this example, the photoelectric conversion unit 11, the image storage unit 12, the barcode detection unit 13, and the barcode scanning unit 14 function as a “scanning data generation unit that generates a plurality of barcode scanning data”.

  For example, the photoelectric conversion unit 11 scans a form on which a barcode is printed, and generates image data of the form. That is, the entire form is scanned, and image data including information other than the barcode is generated. However, it is not always necessary to limit the bar code to that shown on the form (paper).

  The photoelectric conversion unit 11 reads a form two-dimensionally and functions as an “image data generation unit that generates two-dimensional image data including a barcode”. The photoelectric conversion unit 11 includes, for example, photoelectric conversion elements arranged one-dimensionally or two-dimensionally. A form can be read planarly by a two-dimensionally arranged photoelectric conversion element, for example, an imaging camera. In addition, it is possible to read a form in a two-dimensional manner by moving a one-dimensionally arranged photoelectric conversion element, for example, a linear sensor relative to the form (moving one or both of the form and the photoelectric conversion element). it can.

  The image data read by the photoelectric conversion unit 11 is two-dimensional image data in which pixels are arranged in a plane, and the brightness (shading, brightness) of each pixel is represented by a plurality of bits (for example, n bits). Shall. FIG. 2 is a diagram illustrating the form image data acquired by the photoelectric conversion unit 11. Areas A0 and A1 are areas corresponding to the entire form and barcode (barcode areas described later), respectively.

  The image storage unit 12 stores the image data read by the photoelectric conversion unit 11. The image storage unit 12 is composed of, for example, a semiconductor memory and stores image data.

For example, the barcode detection unit 13 detects a barcode region (region where the barcode exists) from the image data stored in the image storage unit 12 as in the following (1) to (4).
(1) Binarization of image The luminance of each pixel is divided into two states, a dark state and a bright state, according to whether or not a predetermined value is exceeded (binarization of luminance of each pixel).
(2) Detection of linear region A region (linear region) in which dark pixels are continuously arranged in a linear shape is detected. This is because black lines (bars) existing in the barcode are extracted. As a result, a plurality of linear regions are detected.
(3) Barcode detection The inclination of each of the plurality of linear regions is calculated, and a linear region having substantially the same inclination is detected as a bar. That is, it is determined that a linear region whose inclination does not coincide with other linear regions does not correspond to a bar.
(4) Barcode area detection The area surrounding the detected bar is detected as a barcode area.

  The barcode scanning unit 14 scans the barcode region a plurality of times while changing the position, and functions as a “scanning data extraction unit that extracts the plurality of scanning data from the generated image data”. FIG. 3 is a schematic diagram showing the result of scanning by the barcode scanning unit 14. Here, the positions (starting points P0 to P4) are changed, scanning is performed approximately parallel five times, and scanning data Li (L0 to L4) is extracted from the image data stored in the image storage unit 12. That is, a series of numerical values representing bar code shading is converted into data for each scan.

  In FIG. 3, the scanning direction is oblique to the bar of the barcode, but may be perpendicular to the bar. It should be noted that by setting the direction of the bar of the bar code obliquely, the data obtained by the scan data L0 to L4 is given diversity, and the high-resolution one-dimensional image reconstructed (generated) by the image reconstruction unit 15 Improves accuracy.

The scan data L0 to L4 are one-dimensional data in which pixels are arranged in a line as shown in the equation (1) (sampling at a predetermined sampling frequency (for example, digital at a predetermined interval on a form) )), And the lightness (darkness, brightness) B of each pixel is represented by a plurality of bits (for example, n bits).
B = Li (k) (1)

In FIG. 3, since the starting points P0 to P4 are arranged in the height direction of the image read by the photoelectric conversion unit 11, a deviation (positional difference Δi) between the starting points P0 to P4 occurs.
On the other hand, “positional difference Δi = 0” can be obtained by changing the position (starting points P0 to P4) along the height direction of the barcode (direction perpendicular to the bar). In scanning, in order to change the position (starting points P0 to P4) in the height direction of the barcode, for example, information on the inclination of the linear region obtained in the process (3) in the barcode detection unit 13 is used. That is, the starting points P0 to P4 are set so that the starting points P0 to P4 of the scanning data L0 to L4 are arranged along this inclination.

  The image reconstruction unit 15 reconstructs one high-resolution scan data from a plurality of scan data (for example, scan data L0 to L4). The image reconstruction unit 15 functions as “a high-resolution processing unit that generates one data series having a resolution higher than the resolution of each of the plurality of scan data based on the plurality of scan data”.

  There is a technique called “SuperResolution” that reconstructs a single high-resolution image from a plurality of low-resolution images with slightly different shooting positions. This is a technique for obtaining a single high-resolution image from, for example, a continuous image obtained by photographing the ground surface from an artificial satellite or a plurality of continuous images of moving images. By using this method, an image including a frequency region exceeding the Nyquist frequency can be reconstructed. On the other hand, for reconstruction, a plurality of images taken by shifting the position little by little are required.

  In principle, it is possible to reconstruct a high-resolution barcode image from a plurality of low-resolution barcode images by super-resolution technology and read a high-density barcode. However, in order to obtain a plurality of images from one form, it is necessary to scan a plurality of times, which takes time.

  Here, the target is limited to one-dimensional barcodes, and one low-resolution image is scanned from multiple one-dimensional images that are scanned multiple times while changing the position in the height direction of the barcode. Reconstruct high-resolution 1D images. At this time, the image reconstruction unit 15 performs the following processes (1) to (3) on the scan data Li.

(1) Estimation of position difference The image reconstruction unit 15 estimates the deviation (position difference Δi) of the starting points P0 to P4 of the scan data L0 to L4. However, the calculation of the position difference Δi can be omitted assuming that the starting points P0 to P4 of the scanning data L0 to L4 are arranged in the height direction of the barcode. In this case, “positional difference Δi = 0”. It is also possible to estimate the position difference Δi based on the angle difference θ between the line segment connecting the starting points P0 to P4 and the height direction of the bar (Δi = Hi * tan (θ), Hi: base point Distance from P0 to base points P1 to P4, i = 1 to 4).

  For example, the image reconstruction unit 15 estimates the positional difference Δi of the scanning data Li (L1 to L4) with the scanning data L0 as a reference. Specifically, Δi that minimizes the following equation (10) is obtained. This can be solved by the linear least square method.

Σ _k {−Δi * (L0 (k + 1) −L0 (k−1)) / 2
+ (Li (k) −L0 (k))} ² ...... Formula (10)

This equation (10) is obtained as follows.
In general, it is considered that the object f (x) is moving at a speed v, and two images f1 (x) and f2 (x) obtained by photographing the object at a minute time interval Δt have the following relationship. .
f2 (x) = f1 (x−v * Δt) Equation (11)

When f (x) is linearly approximated and partially differentiated with respect to x and t, the following equation (12) is established.
v * (∂f (x) / ∂x) + (∂f (x) / ∂t) to 0 (12)

Here, in order to reduce the influence of noise, a value obtained by integrating the square of the left side of Expression (12) is used as an evaluation function, and v and Δt at which this evaluation function is minimized are obtained.
∫ [v * (∂f (x) / ∂x) + (∂f (x) / ∂t)] ² dx
...... Formula (13)

By performing the following replacement in Expression (13) (differentiation and integration are replaced with the difference and addition of digital data), Expression (10) is derived.
∂f / ∂x → (L0 (k + 1) −L0 (k−1)) / 2
∂f / ∂t → Li (k) -L0 (k)
∫dx → Σ _k
v * Δt → −Δi

(2) Wideband Interpolation The image reconstruction unit 15 performs wideband interpolation processing on the scan data Li (L0 to L4). Specifically, for example, the scan data Li is processed by the following equation (20). This wideband interpolation processing corresponds to the processing in “a filter processing unit that performs a filtering process that allows a plurality of scanning data to pass a high-frequency component having a frequency equal to or higher than half the predetermined sampling frequency”.

ei (l) = Σ _k [Li (k) * sin (2 (l / n−k))
/ 2 (l / n−k)] (20)
Where n: Resolution magnification (for example, 10 times)

Hereinafter, the derivation of Expression (20) will be described.
Wideband interpolation means processing by a wideband LPF (low-pass filter) that transmits all of the original signal. The band of the scan data Li or the like is limited to, for example, twice the Nyquist frequency (1/2 of the sampling frequency) (ie, [−2π, 2π]) due to the influence of the sampling frequency. Is done.
In this case, in the frequency space, the skirts of adjacent signal distributions overlap, and the overlap of the aliasing components is double. In other words, the component existing at a certain frequency is the sum of two of the fundamental wave and harmonics.

The input digital data is a data string called Li (k), and what is considered as a continuous function (delta pulse string) for the position (x) corresponds to the sampling signal d (x). A sampled digital signal di (x) is defined by the product of the original signal fi and the δ pulse train s (x).
di (x) = Li (x) s (x)
= Σ _k δ (x−k) Li (k) Equation (21)
s (x) = Σ _k δ (x−k)

The Fourier transform Di (ω) of the sampled digital signal di (x) is expressed by the following equation.
Di (ω) = Fi (ω) * S (ω)
= (1 / 2π) ∫Fi (ω) S (ω−Ω) dΩ
= Σ _k Fi (ω−2πk) (Formula 22)

  First, a low-pass filter LPF is applied to the sampled signal di (x) to extract only the band [−2π, 2π]. Assuming that ei (x) is a signal obtained by subjecting di (x) to LPF processing, its Fourier transform (Ei (ω)) is as follows.

E0 (ω) = LPF (ω) Di (ω)
= F (ω) + LPF (ω) exp (2πiΔi) F (ω + 2π)
+ LPF (ω) exp (−2πiΔi) F (ω−2π)
...... Formula (23)
However,
LPF (ω) = 1 (−2π ≦ ω ≦ 2π)
= 0 (ω <-2π, 2π> ω)

This is a signal obtained by adding half of the fundamental wave and ± 1st harmonic.
The sampling signal di (x) of the scan data Li is expressed as follows.
di (x) = Σ _k δ (x−k) Li (k)

This sampled signal di (x) is multiplied by the broadband LPF of the equation. Since LPF (ω) is a convolution with a sinc function (sin (2x) / (2x)) in real space, a signal ei (x) is obtained.
ei (x) = lpf [di (x)]
= Di (x) * sin (2x) / (2x)
= ∫ [di (x−ξ) * sin (2ξ) / (2ξ)] dξ
...... Formula (24)

  The signal (ei (x)) obtained by this LPF process is a continuous signal, and it is only necessary to calculate equation (24) only for the desired (sampled) position (x). By substituting equation (21) and “x = 1 / n” into equation (24), equation (20) is obtained.

(3) Calculation of Weighted Sum The image reconstruction unit 15 obtains a weighted Wi that satisfies the following expressions (31) and (32) and minimizes the expression (33), and calculates the weighted sum O (k) of the signal ei. calculate. This calculation of the load sum corresponds to the processing in the “adding unit that generates the data series 1 by weighting and adding a plurality of filtered scan data”.

Σ _m Wm = 1 …… Formula (31)
Σ _m exp (2πiΔm) Wk = 0 Equation (32)
Σ _m Wm ² ...... Formula (33)
O (k) = Σ _i Wi * ei (k) (formula (34))

  This load sum O (k) is a data series having a higher resolution than the scan data L0 to L4. That is, the load sum O (k) corresponds to “one data series having a resolution higher than the resolution of each of the plurality of scan data”.

  The decoding unit 16 decodes the reconstructed high-resolution scan data according to the barcode specification.

  The output unit 17 is an output device that outputs a decoding result, for example, a display device (liquid crystal display device) or a printing device.

  In the conventional barcode reader, the scan data for each scan is analyzed and coded by the decoding unit, and multiple scans are performed in case there is some noise in the barcode. , If any of the scan data is correctly decoded, it is output. For this reason, it is necessary that each piece of scan data has a resolution necessary for decoding. For this reason, in order to read a bar code with a module width of 0.167 mm, for example, twice as much as that is necessary. It was necessary to have a resolution of 300 dpi (≈0.084 dots / mm) as the frequency.

  The image reconstruction unit 15 reconstructs one high-resolution scan data from the scan data L0 to L4 by a super-resolution technique. For example, high-resolution scanning data having information exceeding 200 dpi is reconstructed from scanning data obtained by scanning a barcode detected from an image of 200 dpi (≈0.127 dots / mm) a plurality of times while changing the scanning position.

  As a result, even if each piece of scan data does not have the resolution necessary for decoding, high-resolution scan data can be obtained from a plurality of scan data, and a high-density barcode can be obtained from a low-resolution image. Can be read.

(Second Embodiment)
FIG. 4 is a block diagram showing a barcode reader 20 according to the second embodiment of the present invention. The bar code reader 20 is a device that scans a bar code printed on an object with a laser.

  The barcode reader 20 includes a photoelectric conversion unit 21, a barcode scanning unit 24, an image reconstruction unit 15, a decoding unit 16, and an output unit 17.

For example, the photoelectric conversion unit 21 irradiates an object with laser light and converts the luminance of reflected light into an electric signal.
The barcode scanning unit 24 scans the object with the laser light from the photoelectric conversion unit 21 a plurality of times. The position (area) on the object on which the barcode is arranged is specified, and the barcode scanning unit 24 scans the specific area with the laser beam.

  That is, the photoelectric conversion unit 21 and the barcode scanning unit 24 generate scanning data Li (L0 to L4) by scanning the laser beam on the barcode represented on the object a plurality of times. It functions as a scanning data generation unit that generates a plurality of scanning data.

  The bar code reader 20 is omitted from the bar code reader 10 in that the image storage unit 12 and the bar code detector 13 are omitted. By specifying the position on the object on which the barcode is arranged, the image storage unit 12 and the barcode detection unit 13 can be omitted. Other configurations are not substantially different from those of the bar code reader 10 and will not be described.

(Other embodiments)
Embodiments of the present invention are not limited to the above-described embodiments, and can be expanded and modified. The expanded and modified embodiments are also included in the technical scope of the present invention.

DESCRIPTION OF SYMBOLS 10 Barcode reader 11 Photoelectric conversion part 12 Image storage part 13 Barcode detection part 14 Barcode scanning part 15 Image reconstruction part 16 Decoding part 17 Output part

Claims (5)

A scanning data generation unit for generating a plurality of scanning data of the barcode;
A high-resolution processing unit that generates one data series having a resolution higher than the resolution of each of the plurality of scan data based on the plurality of scan data;
A decoding unit for decoding the barcode information from the generated data series;
A bar code reading apparatus comprising: The scan data generation unit
An image data generation unit for generating two-dimensional image data including the barcode;
A scanning data extraction unit that extracts the plurality of scanning data from the generated image data,
The bar code reader according to claim 1. The bar code reader according to claim 1 or 2, wherein the starting points of the plurality of scan data are arranged side by side in the height direction of the bar code. The plurality of scan data is data sampled at a predetermined sampling frequency;
The high-resolution processor is
A filter processing unit that performs a filter process for allowing the plurality of scan data to pass a high-frequency component having a frequency equal to or higher than a half of the predetermined sampling frequency;
An addition unit that generates the one data series by weighted addition of the plurality of scanned data that has been subjected to the filter processing,
The barcode reader according to any one of claims 1 to 3, wherein Generating a plurality of bar code scanning data;
Generating one data series having a resolution higher than the resolution of each of the plurality of scan data based on the plurality of scan data;
Decoding the barcode information from the generated data series;
A bar code reading method comprising:

Images