JP4127187B2 - Code information reading apparatus and program - Google Patents

Description

  The present invention relates to a code information reading apparatus and a program for converting the code of image data read by an imaging means for photographing code information.

  In general, in a code information reading device that reads code information such as a barcode and a two-dimensional code using a charge storage type image sensor (CMOS sensor), the code information recorded on a label or the like is optically captured ( (Hereinafter referred to as “capture”), the shutter time of the CMOS sensor when the image data for one surface is binarized, subjected to processing such as distortion correction, or the necessary portion is cut out and then decoded. That is, in order to optimize the exposure time according to the current environment, the optimum value of the shutter time is calculated based on the image data obtained at the time of the first capture (hereinafter referred to as pre-capture). After setting as a parameter, the image data is captured by restarting the reading operation (hereinafter referred to as the main capture). ), And thereby so as to obtain the image data to be decoded. As a result, the shutter time of the CMOS sensor can be adjusted according to the current environment, and decoding errors can be effectively prevented.

Conventionally, as this type of code information reading apparatus, only the pixel signal thinned out in both the horizontal and vertical directions in the pre-exposure operation of the image sensor is read, and the main exposure operation is performed according to the signal read out in the pre-exposure operation. Is known to read out the signals of all pixels (see Patent Document 1).
JP 2003-132301 A

However, in this type of code information reading apparatus, it is possible to increase the speed and reduce the power by performing a thinning operation at the time of pre-exposure. The control lines HMODE and VMODE to the scanning circuit block and the vertical scanning circuit block are used, and data is thinned out in a fixed manner.
By the way, every time data capture is performed, when transferring all of the image data for one surface to the memory for decoding, the amount of data becomes extremely large when using a high-resolution image sensor, As a result, the data transfer time increases and the time until the decoding process starts increases.

  An object of the present invention is to enable transfer of only the minimum necessary data in the process of transferring image data read by an imaging means for capturing code information to a memory, and to provide the minimum necessary data. However, it is to ensure that the code conversion can be executed.

The invention according to claim 1 is a code information reading device that converts image data read by an image pickup unit that captures code information, and sets a decimation number for thinning out pixels constituting the image data. Means, a thinning means for thinning out the image data based on the thinning number set by the setting means, a transfer means for transferring the image data thinned out by the thinning means to a memory, and an image in the memory A conversion unit that decodes the data and performs code conversion; and if the conversion unit recognizes that normal decoding of the image data is impossible, the new decimation number is reduced, By recursively repeating the process of decoding the image data based on the thinning number until it is decoded normally Optimum decimation number determined, characterized by comprising a resetting means for resetting the optimum decimation number and Te.
Furthermore, a program for causing a computer to function as the main means shown in the above-described invention according to claim 1 is provided (the invention according to claim 2).

  According to the first aspect of the present invention, in the state in which a thinning number for thinning out the pixels constituting the image data read by the imaging means for photographing the code information is set, the image data is obtained based on the thinning number. If normal decoding is not possible when the thinned image data is decoded and code conversion is performed after thinning processing is performed, this new thinning is performed after reducing the currently set thinning number. Since the process of decoding while thinning out the image data based on the number is recursively repeated until it is normally decoded, the optimum thinning number is determined, and the optimum thinning number is reset. In the process up to transferring the image data read by the image pickup means to the memory, only the minimum necessary data can be transferred. With shortening the data transfer time, even in the minimum amount of data necessary, it is possible to perform the transcoding reliably.

(Example 1)
Hereinafter, a first embodiment of the present invention will be described with reference to FIGS.
FIG. 1 is a block diagram showing the overall configuration of the code information reading apparatus in this embodiment.
The code information reading apparatus includes a general-purpose CMOS sensor unit 10, an illumination LED 20, an illumination LED 30, a CMOS sensor / LED controller 40, a CPU 50, and a memory 60 as basic components. In the figure, circled numbers are sequence numbers indicating the flow of processing, and each process is sequentially executed according to the circled numbers (the same applies hereinafter).

  The general-purpose CMOS sensor unit 10 is a light-receiving element array (area sensor) that captures reflected light by optically scanning a label or the like on which code information to be read (two-dimensional code information) is recorded by the illumination LED 20 and the illumination LED 30. ) 11, a driver 12 for driving the area sensor 11, an address buffer 13 for designating an area to be read by the area sensor 11, and a control & control for controlling the entire general-purpose CMOS sensor unit 10 It has a logic unit 14 and an AD converter 15 for analog / digital conversion of the received amount (charge amount).

The CMOS sensor / LED controller 40 controls the general-purpose CMOS sensor unit 10, the illumination LED 20, and the irradiation LED 30 under the control of the CPU 50, and a data buffer that captures output data from the controller unit 41 and the AD converter 15. 42 and an AE integrating circuit 43.
The AE integration circuit 43 calculates an optimum value of the shutter time by obtaining a data integration value for one surface based on image data (output data from the AD converter 15) obtained at the time of pre-capture. After this optimum value is set as a parameter, the image data is fully captured by restarting the reading operation. In other words, in order to optimize the shutter time (exposure time) according to the environment, the optimum value of the shutter time is calculated based on the image data obtained at the time of pre-capture, and this optimum value is set as a parameter before reading. The image data is fully captured by restarting the operation, and the image data after the actual capture is binarized or subjected to processing such as distortion correction and then decoded.

  The memory 60 is a RAM (for example, a static RAM) or the like, and has a program execution area and a work area. In this work area, a thinning number K in the column direction and a thinning number L in the row direction are set. . The matrix thinning-out numbers K and L are thinning-out numbers for thinning out image data for one surface at equal intervals in the column direction and the row direction when image data is output from the general-purpose CMOS sensor unit 10. Although the default is arbitrarily set in advance by the user in advance, one of matrix thinning-out numbers K and L may be set. In this embodiment, the matrix thinning-out numbers K and L are not fixed data, but are variable data that is changed based on whether or not the decoding can be performed normally.

  That is, after data thinning is performed based on the matrix thinning numbers K and L set as default, when the image data after the thinning is decoded, it is checked whether or not the decoding can be normally performed. When decoding is successful, the matrix thinning numbers K and L are not changed. However, when a decoding error occurs, the current matrix thinning numbers K and L are gradually reduced. After that, the decoding process while thinning out the image data based on the new thinning number is repeatedly executed until it is normally decoded, and when the decoding is successful, the thinning number is determined as the optimum thinning number. The thinning numbers K and L are set in the memory 60 again. Thereafter, data is thinned out using the optimum thinning-out numbers K and L.

  When outputting image data from the general-purpose CMOS sensor unit 10, that is, when image data for one surface is output from the AD converter 15, the image data for one surface is converted based on the matrix thinning-out numbers K and L. A process for thinning out data at equal intervals in the column direction and the row direction is executed. At this time, for example, “2” is set as the column thinning number K and “2” is set as the row thinning number L. In this case, the image data of the matrix configuration is thinned out in units of two pixels in both the column direction and the row direction, so that the original image data for one surface is reduced to 1/9 of the data amount. .

FIG. 2 is a diagram for specifically explaining an operation when data is thinned out at equal intervals in the column direction and the row direction when image data for one surface is output from the general-purpose CMOS sensor unit 10. .
In this case, the image data for one surface output from the general-purpose CMOS sensor unit 10 has a configuration of 3n rows and 3m columns in the illustrated example, and each data in the first row includes “A1, 1”, “A1, 2”,..., “A1, K + 1”,... “A1, 3m”, and the data in the second row are “A2, 1”, “A2, 2”,. “A2, K + 1”,... “A2, 3m”, and the data in the 3n-th row are “A3n, 1”, “A3n, 2”,..., “A3n, K + 1”,. A3n, 3m ".

  Here, when the column thinning number K is set to “2” and the row thinning number L is set to “2”, thinning is performed every K (two) in the column direction of the image data. Therefore, the data in the first row are “A1, 1”, “A1, 4”,... “A1, 3m-2”. Similarly, in the column direction of the image data, thinning is performed every L (two), so the next row data following the first row is “A4, 1”, “A4” on the fourth row. 4 ”,...“ A4, 3m-2 ”, and the data on the 3n-2nd line are“ A3n-2, 1 ”,“ A3n-2, 4 ”,. "

  Next, the operation algorithm of the code information reading apparatus in the first embodiment will be described with reference to the flowchart shown in FIG. Here, each function described in the flowchart is stored in the form of a readable program code, and operations according to the program code are sequentially executed. It is also possible to sequentially execute operations according to the program code transmitted via the transmission medium. The same applies to other embodiments to be described later. In addition to a recording medium, an operation peculiar to this embodiment can be executed using a program / data supplied externally via a transmission medium.

FIG. 3 is a flowchart showing a decoding sequence until the code information is read and decoded, and FIG. 4 is a diagram showing data output timing. In the drawing, circled numbers are sequence numbers indicating the flow of each process, as described above, and each process is sequentially executed according to the circled numbers.
First, the CPU 50 requests initialization of the general-purpose CMOS sensor unit 10 via the CMOS sensor / LED controller 40 (step A1), and after that, the column thinning number K and the row are thinned as transfer data thinning numbers from the memory 60. The thinning-out number L is read (step A2). In this state, a pre-capture command for one surface is performed (step A3). Each command data (system data) is output in synchronization with the system CLK (clock) signal shown in FIG. 4A (see FIG. 4B, the same applies hereinafter).

 Thus, the image data for one surface obtained from the general-purpose CMOS sensor unit 10 is captured (step A4), and when this image data is transferred to the data buffer 42 in the CMOS sensor / LED controller 40, After performing data thinning according to the thinning number K and the thinning number L, the image data in the data buffer 42 is transferred to the data buffer 42 and adjusted to the data bus size between the data buffer 42 and the memory 60. Are bus-sized for each row unit and sequentially transferred to the memory 60 (step A5). The ADC data (input data to the data buffer 42) output from the AD converter 15 in accordance with the master CLK signal (ADC data synchronization signal) shown in FIG. 4C is as shown in FIG. If the data transfer amount during conventional pre-capture is (1 / K + 1) × (1 / L + 1) times and the matrix decimation numbers K and L are “2”, the image data after decimation is 1 Therefore, the data transfer time is reduced by the AT time.

  Based on the thinned image data obtained at the time of pre-capturing in this way, the AE integrating circuit 43 obtains the data integrated value and calculates the optimum value of the shutter time (step A6), and then the CMOS sensor / LED controller. 40 sets the optimum value as a parameter in the general-purpose CMOS sensor unit 10 (step A7). In this state, the CPU 50 issues a main capture command for one surface to the general-purpose CMOS sensor unit 10 via the CMOS sensor / LED controller 40 (step A8). As a result, the image data obtained again from the general-purpose CMOS sensor unit 10 is fully captured (step A9) and transferred to the memory 60 via the data buffer 42 in the CMOS sensor / LED controller 40. As in the pre-capture, after data thinning is performed according to the column thinning number K and the row thinning number L, the data is transferred to the data buffer 42 and the image data in the data buffer 42 is bus-sized for each row. Then, the data are sequentially transferred to the memory 60 (Step A10). Here, as shown in FIG. 4D, the image data after the main capture is also (1 / K + 1) × (1 / L + 1) times larger than the data transfer amount at the time of the conventional main capture.

  In this embodiment, in order to increase the accuracy of the reading operation, the main capture is repeated a plurality of times (steps A8 to A11). Each image data obtained for each main capture is obtained each time. Data thinning is performed according to column thinning number K and row thinning number L, and data comparison processing is executed to detect the presence or absence of the matching based on the thinned image data (step A11). Here, on the condition that the coincidence of each data is detected (YES in step A11), the data processing is started (step A12). That is, after binarizing the image data and performing distortion correction or the like, the code information is further cut out. Then, a decoding process is executed based on the image data after the data processing (step A13), and it is checked whether or not the decoding was successful (step A14).

  If a decoding error has occurred (NO in step A14), a process of subtracting “1” from the column thinning number K and the row thinning number L is performed to change the value (step A15). Subsequently, the above-described operation is repeated while performing pre-capture based on the thinned-out numbers K and L after this change (steps A2 to A14). As a result, if the decoding can be normally performed (YES in step A14), the thinning numbers K and L at that time are determined as the optimum thinning numbers, and thereafter, the optimum thinning numbers K and L are fixed. Decimate data. In this case, the capture data after being fixed to the optimum thinning-out numbers K and L is, for example, as shown in FIG. 4D, and a considerable amount of data is reduced compared to the data transfer amount at the time of conventional main capture. Therefore, the data transfer time is shortened by BT.

  As described above, the code information reading apparatus according to the first embodiment uses the thinned-out number of the image data read by the general-purpose CMOS sensor unit 10 in the state where the thinned-out numbers K and L for thinning out the image data are set. If a decoding error occurs when the thinned image data is decoded after performing thinning processing based on K and L, the currently set thinning numbers K and L are reduced to one level. Since the optimum thinning number is determined by repeatedly executing the decoding process while thinning out the image data based on the new thinning number until normal decoding is performed, the optimum thinning number is reset. In the process until the image data read by the CMOS sensor unit 10 is transferred to the memory 60, only the necessary minimum data is obtained. Can be sent, together with the shortening of the data transfer time, even in the minimum amount of data necessary, it is possible to perform the decoding reliably.

In this case, when the image data is output from the general-purpose CMOS sensor unit 10, the process of thinning out the image data is started, so that the image data output from the general-purpose CMOS sensor unit 10 is written into the data buffer 42. Data thinning can be performed before, and the data transfer time can be efficiently shortened.
Further, the thinning numbers K and L are set corresponding to the column direction and the row direction of the image data for one surface, and the image data are equally spaced based on the thinning number K and the thinning number L in the row direction. Therefore, average thinning is possible as a whole, and adverse effects due to thinning can be suppressed as much as possible when decoding image data.

(Second Embodiment)
Hereinafter, a second embodiment of the present invention will be described with reference to FIGS. In the first embodiment described above, the processing for thinning out the image data is started when the image data is output from the general-purpose CMOS sensor unit 10. In the second embodiment, the general-purpose CMOS sensor is used. After the image data output from the unit 10 is written in the data buffer 42, the image data is thinned out when the image data is transferred from the data buffer 42 to the memory 60.
Here, components that are basically the same in both embodiments are denoted by the same reference numerals, description thereof will be omitted, and the following description will focus on the features of the second embodiment.

FIG. 5 specifically shows how data thinning is performed when image data is transferred from the data buffer 42 to the memory 60 after the image data output from the general-purpose CMOS sensor unit 10 is written to the data buffer 42. It is the figure shown in.
FIG. 5A shows image data for one surface output from the general-purpose CMOS sensor unit 10, and the image data for one surface is written in the data buffer 42 as it is (see FIG. 5B). In this example, the column thinning number K is set to “2” and the row thinning number L is set to “2” as in the first embodiment described above.

  In this case, since the data in the first row becomes the target data to be transferred to the memory 60, all the data is sent to the work transfer area in the CMOS sensor / LED controller 40 (FIG. 5C In this work transfer area, when a column to be transferred to the memory 60 is selected, as shown in FIG. 5D, the data in the first row is “A1, 1”. , “A1, 4”,... “A1, 3m-2”. Thereafter, a row that is not subject to thinning is read out as a transfer subject row, and a column that is not subject to thinning is selected and transferred to the memory 60 from within this subject row. By performing such a thinning process, the data finally transferred from the data buffer 42 to the memory 60 is as shown in FIG. 5E, and “A1, 1”, “A1” in the first row , “4”,... “A1, 3m-2”, “A3n-2, 1”, “A3n-2, 4” in the A3n row,. As a result, the image data after thinning becomes 1/9 of the data amount of the original data for one side.

FIG. 6 is a flowchart showing a decoding sequence until code information is read and decoded in the second embodiment, and FIG. 7 is a diagram showing data output timing.
First, the CPU 50 requests initialization of the general-purpose CMOS sensor unit 10 via the CMOS sensor / LED controller 40 (step B1), and after that, the column thinning number K and the row are thinned as transfer data thinning numbers from the memory 60. The thinning-out number L is read (step B2). In this state, a pre-capture command for one surface is performed (step B3). Each command data (system data) is output in synchronization with the system CLK (clock) signal shown in FIG. 7A (see FIG. 7B, the same applies hereinafter).

Thereby, the image data for one surface obtained from the general-purpose CMOS sensor unit 10 is captured (step B4), and the image data for one surface is transferred as it is to the data buffer 42 in the CMOS sensor / LED controller 40. Later, data thinning is performed according to the column thinning number K and the row thinning number L, and is transferred to the memory 60 (step B5).
In this case, the ADC data output from the AD converter 15 in synchronization with the master CLK signal shown in FIG. 7C is as shown in FIG. Further, when data is transferred from the data buffer 42 in the CMOS sensor / LED controller 40 to the memory 60, the data is divided and transferred in units of rows. At this time, as shown in FIG. In the second embodiment, the data thinned out in accordance with the column thinning number K is divided and transferred, whereas in the second embodiment, all the data for one row is sequentially transferred divided into row units. As shown in FIG. 7 (F), the amount of data per row can be greatly reduced, and when data is transferred in units of rows, data thinning according to the thinning number L is performed. Therefore, the number of divisions can be greatly reduced.

 When the image data for one surface pre-captured in this way is transferred to the data buffer 42, the AE integrating circuit 43 obtains the data integrated value and calculates the optimum value for the shutter time (step B6). ), The CMOS sensor / LED controller 40 sets the optimum value as a parameter in the general-purpose CMOS sensor unit 10 (step B7). In this state, the CPU 50 issues a main capture command for one surface to the general-purpose CMOS sensor unit 10 via the CMOS sensor / LED controller 40 (step B8). As a result, the image data obtained again from the general-purpose CMOS sensor unit 10 is fully captured (step B9).

 The image data for one surface is transferred as it is to the data buffer 42 in the CMOS sensor / LED controller 40 (step B10), and then the image data in the data buffer 42 is transferred to the memory 60. Similarly to the pre-capture described above, data thinning is performed according to the column thinning number K and the row thinning number L (step B11). Even in this case, as shown in FIG. 7F, the amount of data per row can be greatly reduced, and the number of divisions can be greatly reduced. Thereafter, in order to increase the accuracy of the reading operation, the main capture is repeated a plurality of times, and a data comparison process for detecting the match based on each image data obtained for each main capture is executed (steps B8 to B12). ). Even in this case, each image data obtained for each main capture is thinned in accordance with the column thinning number K and the row thinning number L each time, and whether or not they match is determined based on each image data after thinning. Detect (step B12).

  Here, on condition that a match of each data is detected (YES in step B12), the image data is binarized, and after performing distortion correction and the like, further data processing for extracting code information Is started (step B13). Then, a decoding process is executed based on the image data after the data processing (step B14), and it is checked whether or not the decoding was successful (step B15). If a decoding error occurs (NO in step B15), a process of subtracting “1” from the column thinning number K and the row thinning number L is performed to change the value (step B16).

  Then, in a state where the image data for one surface transferred to the data buffer 42 is held as it is, the process returns to Step B11, and the image data in the data buffer 42 is thinned according to the column thinning number K and the row thinning number L below. The above operation is repeated (steps B11 to B16). As a result, if the decoding can be performed normally (YES in step B15), the thinning-out numbers K and L at that time are determined as the optimum thinning-out numbers, and thereafter, the optimum thinning-out numbers K and L are fixed to these. Decimate data. In this case, the capture data after being fixed to the optimum thinning-out numbers K and L is, for example, as shown in FIG. 4 (F), and a considerable amount of data is reduced compared with the data transfer amount at the time of conventional main capture. be able to.

  As described above, the code information reading apparatus according to the second embodiment has an image for one surface read by the general-purpose CMOS sensor unit 10 in a state where the thinning numbers K and L for thinning out image data are set. After transferring the data to the data buffer 42 in the controller unit 41 as it is, when transferring the image data from the data buffer 42 to the memory 60, a thinning process is performed based on the thinning numbers K and L, and the image after the thinning is performed. If a decoding error occurs when data is decoded, the currently set thinning-out numbers K and L are reduced to one level, and then decoding is performed while thinning out image data based on the new thinning-out number To determine the optimum thinning-out number by repeatedly executing until it is successfully decoded, and reset this optimum thinning-out number Therefore, in the process until the image data read by the general-purpose CMOS sensor unit 10 is transferred to the memory 60, only the minimum necessary data can be transferred. Data can be reliably decoded, and when the optimum thinning number is reset, decoding can be repeatedly executed based on the image data held in the data buffer. Need not be repeated over and over, and the data transfer time can be further shortened.

In the first and second embodiments described above, a case is shown in which the default is set to “2” as the column thinning number K and “2” as the row thinning number L. It is optional depending on the high resolution of the image data.
In addition, the code information reading device of each embodiment is not limited to the stand-alone type, and each component is physically separated into two or more casings, a wired transmission path such as a communication line or a cable, or a radio wave, You may comprise the distributed computer system which transmits / receives data via wireless transmission paths, such as a microwave and infrared rays.

  On the other hand, you may make it provide the recording medium which each recorded the program code for making a computer perform each means mentioned above. That is, a recording medium having a program code readable by a computer, a function for setting a thinning number for thinning out pixels constituting image data read by an imaging means for photographing code information, and a set thinning A function of thinning out the image data based on the number, a function of transferring the thinned image data to the memory, a function of decoding the image data in the memory and performing code conversion, and normal decoding of the image data When it is recognized that the image data cannot be read, after the set thinning number is reduced, the process of decoding the image data while thinning out based on the new thinning number is recursively performed until normal decoding is performed. The process for determining the optimal decimation number by repeatedly executing this function and realizing the function of resetting this optimal decimation number It may be a computer which records a gram to provide a recording medium readable.

The block diagram which showed the whole structure of the code information reader. The figure for demonstrating concretely the operation | movement at the time of thinning out data at equal intervals in the column direction and row direction, when the image data for one surface is output from the general purpose CMOS sensor part 10. FIG. The flowchart which showed the decoding sequence until it reads and decodes code information. The figure which showed the output timing of data. In the second embodiment, after image data output from the general-purpose CMOS sensor unit 10 is written in the data buffer 42, when image data is transferred from the data buffer 42 to the memory 60, the state of data thinning is performed. The figure shown concretely. The flowchart which showed the decoding sequence until it reads and decodes code information in 2nd Example. The figure which showed the output timing of the data in 2nd Example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 General-purpose CMOS sensor part 11 Area sensor 12 Driver 14 Control & logic part 15 AD converter 20 LED for illumination
30 LED for irradiation
40 CMOS sensor / LED controller 41 Controller section 42 Data buffer 42
50 CPU
60 memory

Claims (2)

A code information reading device that converts image data read by an imaging unit that captures code information,
Setting means for setting a thinning number for thinning out pixels constituting the image data;
Thinning means for performing a process of thinning the image data based on the thinning number set by the setting means;
Transfer means for transferring the image data thinned out by the thinning means to a memory;
Conversion means for decoding the image data in the memory and performing code conversion;
If the conversion means recognizes that normal decoding of the image data is impossible, the image data is decoded while being thinned out based on the new thinning number after reducing the set thinning number. Resetting means for determining the optimum thinning number by recursively repetitively executing the process until normal decoding is performed, and resetting the optimum thinning number ;
Code information reading apparatus characterized by comprising a. The computer,
Means for setting a thinning-out number for thinning out pixels constituting the image data read by the image pickup means for photographing the code information,
Means for thinning out the image data based on a set thinning number;
Means for transferring the thinned image data to a memory;
Means for decoding the image data in the memory and performing code conversion;
When it is recognized that normal decoding of the image data is impossible, after the set number of thinnings is reduced, normal decoding is performed while decoding the image data based on the new thinning number Means for determining the optimum thinning-out number by recursively executing until it is reconfigured ,
Program to function as.

Images