JP4143044B2 - Reader - Google Patents

Description

  The present invention relates to a reading device that reads information displayed on a medium such as a driver's license or a card. More specifically, the present invention relates to a reading apparatus particularly suitable for reading a watermarked medium.

  Unmanned card issuing machines that issue tickets, credit cards, and the like are widespread. In this type of card issuing machine, in order for the operator at a remote place to confirm the identity of the user or to save the user from entering the name and address, the user's license, for example, is taken and imaged there. There are some equipped with a reading device that reads the displayed face photo and character information. For example, in a terminal that issues tickets and credit cards unattended, a system is conceivable in which a driver's license is inserted into the terminal as personal authentication, and this is imaged by a reader and transmitted to an operator at a remote location. In this system, the operator compares the face photograph of the read image with the face reflected on the surveillance camera and collates them, and can issue a ticket only when it is determined that they are the same person.

  Further, since such a reading device simply images the driver's license, there is a possibility that even if the driver's license is forged using a copy machine or the like, it is read as it is and issued to an unauthorized person. .

  On the other hand, a watermark is formed on the driver's license. For this reason, in order to confirm the authenticity of the driver's license taken in by the card issuing machine, it is conceivable to detect a watermark by a reading device. As a method for detecting the watermark, a method of illuminating the entire surface on both sides of the driver's license and photographing from the front side (Japanese Patent Laid-Open No. 2000-11147), or illuminating the entire surface of the back of the driver's license from the front side. A method of photographing (Japanese Patent Laid-Open No. 11-66273) is known. By adopting these methods, it is possible to issue an illegal ticketing by detecting the watermark of the driver's license captured in the card issuing machine and confirming that it is a genuine driver's license. Can be prevented.

JP 2000-11147 JP-A-11-66273

  However, in the above-described reading device, when a method of illuminating and photographing both the front and back sides of the driver's license is adopted, a watermark pattern overlaps with the character portion of the driver's license, and the character recognition rate decreases. . In order to prevent this, it is conceivable to illuminate only from the front side to perform character recognition and then illuminate from the back side to perform watermark recognition. However, this requires two scans for reading a single license, which takes a long time for reading processing and data transfer, and impedes quick ticketing. On the other hand, when the method of illuminating and photographing the entire back side of the driver's license is adopted, the S / N ratio of the character is lowered and the character recognition rate is lowered because the surface is not illuminated.

  In addition, for the character area and the face photograph area, gradation correction of an image, that is, γ correction is determined from the viewpoint of performing character recognition by image processing or being easy to see when displayed on a display. However, when reading the watermark pattern, the contrast of the watermark pattern is lower than that of characters or photos written on the surface. Difficult to see on the display.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a reading apparatus that can quickly and accurately capture an image while detecting the presence or absence of a watermark on a medium such as a driver's license. It is another object of the present invention to provide a reading device that is easy to visually recognize on a display even when reading a watermark pattern, not limited to image processing and character recognition in a character area and a face photograph area.

In order to achieve the above object, a reading apparatus according to claim 1 includes a first light source for irradiating information recorded on the surface of the medium from the surface side of the medium, and a watermark provided on the medium. a second light source for reading the information recorded on the surface of the watermark reading section in a and the medium reading a pattern, morphism irradiation from the back side of the medium in the watermark region is set in a range that does not affect to recognize When, along with reading the information recorded on the surface of the medium, in a state in which light is irradiated from the second light source, a reading means for reading an image of the watermark, outside of the watermark reading section A first shading correction table for use in reading information recorded on the surface of the medium; information recorded on the medium surface in the watermark reading section; A second shading correction table used for reading a pattern image, and the second shading correction table has a portion corresponding to the watermark pattern area of the first shading correction table as the watermark pattern. When the watermark reading section is detected, the first shading correction table is switched to the second shading correction table .

Therefore, since the second light source irradiates only the watermark pattern area of the medium, it is possible to prevent the watermark from overlapping the information recorded on the surface of the medium. Therefore, it is possible to perform it is possible to read both the record information and the watermark in a single scan, the reading of the recorded information with high accuracy. Moreover, when reading in the watermark reading section including the watermark pattern area, the watermark pattern can be read with high accuracy while maintaining high reading accuracy of the character portion.

According to a second aspect of the present invention, in the reading device according to the first aspect, the first γ correction curve used when reading information recorded on the surface of the medium outside the watermark reading section, and the watermark reading A correction table recording information recorded on the medium surface in the section and a second γ correction curve used for reading the watermark pattern image, and the first γ correction curve when the watermark reading section is detected To the second γ correction curve .

Therefore, the watermark pattern is emphasized only when the watermark pattern area is read and displayed as an image that can be easily viewed on the display. Therefore, the watermark pattern can be read with high accuracy while maintaining high reading accuracy of the character portion. It can be carried out.

  Further, in the invention described in claim 3, in the reading device described in claim 1 or 2, the reading means is a line sensor capable of reading image data for at least one line of the medium. Therefore, a small and inexpensive reading device can be obtained using a general-purpose line sensor.

Further, the invention of claim 4, wherein the second shading correction table definitive the reading device according to claim 1 Symbol placement is partial amplitude of the correction data in the range of the watermark pattern area based on the first shading correction table The image gain of the watermark pattern area is lowered to be lower than the image gain outside the range of the watermark pattern area . In this case, it is possible to obtain a good image having a higher S / N ratio than the data obtained by learning by actually turning on the second light source.

As described above, according to the reading device of the first aspect, since both the recorded information and the watermark can be read in one scan, the processing time can be reduced while reading the recorded information with high accuracy. Shortening and increase of traffic at the time of transfer to a remote place can be prevented. Further, when a watermark pattern image is read, the watermark portion can be read with high accuracy while maintaining high reading accuracy of the character portion. In addition, since a plurality of shading correction tables are switched and used, it becomes easy to switch the correction amount at the same time that the second light source is turned on at an arbitrary timing, and the weight of the system can be reduced.

According to the reading device of the second aspect , the watermark pattern is emphasized when the watermark pattern area is read and displayed as an image that can be easily viewed on the display, so that the watermark pattern can be easily confirmed.

  Furthermore, according to the reading device of the third aspect, a small and inexpensive reading device can be obtained using a general-purpose line sensor.

Further, according to the reading device of the fourth aspect, it is possible to obtain a good image having a higher S / N ratio than the data obtained by learning by actually turning on the second light source.

  Hereinafter, the configuration of the present invention will be described in detail based on the best mode shown in the drawings.

1 and 2 show an example of an embodiment of a reading device 1 of the present invention. The reading device 1 includes a first light source 3 for irradiating information recorded on the surface of the medium 2 from the surface side of the medium 2 and only a watermark pattern area 9 provided on the medium 2. The second light source 5 that emits light from the back side of the medium and the information recorded on the surface of the medium 2 are read, and the image of the watermark pattern is read in the state where the light is emitted from the second light source 5. The reading means 6 is provided.

  In the present embodiment, a driver's license is used as the medium 2. As shown in FIG. 3A, the license has a character portion 7 indicating personal information such as an address and name on the upper side on the front side and a face photograph portion 8 on the right side. Since the character portion 7 recognizes characters and the face photograph portion 8 is visually confirmed, it is necessary to avoid deterioration of the S / N ratio caused by watermarks or transmission spots of the medium material due to transmitted light. Here, since the column such as the license condition located in the center on the left side is often unnecessary for personal authentication, the watermark is projected as the watermark pattern area 9 (indicated by hatching in FIG. 3A). To do.

  The reading unit 6 is a line sensor that can read image data for at least one line of the medium 2. Further, the reading unit 6 is a contact sensor and includes the first light source 3. Therefore, it is possible to perform illumination in close contact with the medium 2 and to improve reading accuracy.

  The second light source 5 is disposed on the side opposite to the first light source 3 with respect to the medium 2. The second light source 5 has an arbitrary irradiation range 10 by using a light shielding plate or the like, and illuminates only the watermark pattern region 9 from which a watermark image is to be obtained. As each light source 3 and 5, white LED is used. Then, the density of the watermark in the image is appropriately set by adjusting the irradiation ratio of the light sources 3 and 5.

  In addition, the reading device 1 emits a watermark pattern image by irradiating light from the first shading correction table 11 used when reading information recorded on the surface of the medium 2 and the second light source 5. A second shading correction table 12 used for reading is provided. The first shading correction table 11 is used when only the first light source 3 emits light, and the second shading correction table 12 is used when the first and second light sources 3 and 5 emit light.

  The shading correction tables 11 and 12 are learned at the time of factory shipment, when the apparatus is installed on the site, or during periodic maintenance, and are stored in a memory (not shown). This learning is called shading learning, and processing is performed in which the entire white card faces the reading means 6 and the image output at that time is stored in the memory as a reference level data table. When storing in the memory, noise is removed and white color development is suppressed to prevent whiteout of the image. In addition, data processing may be performed as necessary, such as replacement with an inverse number so that the correction calculation can be completed easily during correction by table reference.

  The second shading correction table 12 increases the amplitude of the correction data only in the watermark pattern region 9, that is, the irradiation range of the second light source 5 (indicated by reference numeral 12a in FIG. 3B), and as a result, the image gain decreases. Like that. As a method of increasing the amplitude, an arbitrary method such as multiplying the correction value of the first shading correction table 11 by, for example, a uniform magnification of 1.5 times is used. For this reason, by changing the shading correction amount of the watermark pattern area 9 partially from the normal time, good image quality similar to that in the normal time is ensured in the area other than the watermark pattern area 9, and also in the watermark pattern area 9. It is possible to obtain image density and contrast that make it easy to recognize a watermark.

  Here, since the second shading correction table 12 is created on the basis of the first shading correction table 11, S / is more than the data obtained by learning by actually turning on the second light source 5. A good image with a high N ratio can be obtained. This is because when the second light source 5 is actually turned on and learning, the transmitted light spots due to the material of the medium 2 become intense.

  Also, the amplitude is increased in order to prevent the transmission of the watermark pattern region 9 from becoming too large due to the addition of the transmitted illumination to the illumination by normal reading, resulting in a whiteout and a watermark image not being obtained. It is conceivable to darken the first light source 3 or reduce the overall image gain so that the watermark image does not fly out. However, if the watermark pattern is set to appear, the character portion 7 and the face photo portion 8 become dark images. This is not preferable because character recognition and face authentication are hindered. Note that the density of the watermark pattern region 9 can be changed by appropriately changing the correction amount of the second shading correction table 12.

  Further, when the second shading correction table 12 is created, normal image reduction learning, that is, generation from the surface image data of the white medium 2 reduces image degradation caused by the brightness spots of the medium 2 due to transmitted light. Yes.

  On the other hand, the reading device 1 includes a roller 13 for transporting the medium 2 and a control board 14 for controlling the position of the medium 2 and each light source 3, 5 and the reading means 4. The control board 14 is further connected to the host device 15. The control board 14 includes a sampling unit 16 that reads an image, a shading correction unit 17 that performs correction by selecting a shading correction table, a γ correction unit 18 that performs γ correction, and an image data output unit 19 that outputs image data. It has.

Note that the γ correction unit 18 performs gradation correction of an image, that is, γ correction (contrast adjustment) so that character recognition by image processing and display on a display are easy to see. This γ correction may be performed using a single correction data, that is, a γ correction curve. However, when reading a watermark pattern, since the contrast of the watermark pattern is lower than that of characters or photographs written on the surface, The same gamma correction as in the face photo area makes it difficult to see on the display. Therefore, similarly to the above-described switching of the shading correction table, a first γ correction curve used in the character area and the face photograph area and a second γ correction curve used in the watermark reading section 21 are provided, and these are set according to the reading area. It is preferable to perform γ correction appropriately by switching. The switching timing between the first γ correction curve and the second γ correction curve is the same as the switching of the above-described shading correction table, for example. Specifically, a correction table in which the first γ correction curve and the second γ correction curve are recorded in the memory in advance is created, and the correction table used in the watermark reading section 21 and other areas is created. Switch. For example, in the case of the first γ correction curve used in a character region or a face photograph region where contrast is clear to a certain extent, the curve is as shown in FIG. 7A, but the second γ used in the watermark reading section 21 is used. correction curve has become any correction curve visually easily contrast on the display watermark is highlighted in the watermark region 9 is obtained as shown in (B) of FIG. The second γ correction curve shown in FIG. 7 (B) is an example to match the case where the concentration of the watermark in the watermark region 9 are distributed in the white side, in white the intermediate density side region Correction for enhancing the contrast is performed.

  The operation of the reading device 1 described above will be described based on the flowchart shown in FIG. 6 and the time chart shown in FIG.

  When it is detected that the medium 2 is taken in from the gate 20, the roller 13 is driven and the medium 2 is conveyed (step 1 (S1)). The first light source 3 is turned on immediately before the medium 2 is applied to the reading means 6 (step 2 (S2)). As the shading correction data, the first shading correction table 11 is used.

  It is determined whether the medium 2 has reached the imaging section (step 3 (S3)). If not reached (step 3 (S3); No), it is judged again. If it has reached (step 3 (S3); Yes), imaging is started (step 4 (S4)).

  Imaging is performed by the reading means 6 (step 5 (S5)). Then, it is determined whether the watermark reading section 21 has been entered (step 6 (S6)). If not (step 6 (S6); No), the imaging is continued (step 5 (S5)). If it is present (step 6 (S6); Yes), the second light source 5 is turned on (step 7 (S7)). At the same time, the shading correction data is switched to the second shading correction table 12 (step 8 (S8)). Note that instead of the shading correction data, γ correction data may be used to switch to the second γ correction curve, or both may be performed.

  Further, imaging is continued (step 9 (S9)), and it is determined whether or not the watermark reading section 21 has been exited (step 10 (S10)). If not (step 10 (S10); Yes), the imaging is continued (step 9 (S9)). If it has come out (step 10 (S10); No), the second light source 5 is turned off (step 11 (S11)). At the same time, the shading data is switched to the first shading correction table 11 (step 12 (S12)). Note that instead of the shading correction data, γ correction data may be used to switch to the first γ correction curve, or both may be performed.

Further, the imaging is continued (step 13 (S13)), and it is determined whether the imaging section has been exited (step 14 (S14)). If not (step 14 (S14); Yes), the imaging is continued (step 13 (S13)). If it has come out (step 14 (S14); No), imaging is completed (step 15 (S15)), and the first light source 3 is turned off (step 16 (S16)). The read data after the shading correction is subjected to processing such as γ correction and output to a display or the like for image display. In the γ correction, it is preferable to switch between the first γ correction curve used in the character area and the face photograph area and the second γ correction curve used in the watermark reading section 21 according to the reading area. For example, when only a watermark pattern or characters or images are required, it is possible to output after performing either γ correction.

  The above-described embodiment is an example of a preferred embodiment of the present invention, but is not limited thereto, and various modifications can be made without departing from the scope of the present invention. For example, in the present embodiment, a driver's license is used as the medium 2. However, the present invention is not limited to this, and any pattern may be used as long as the medium 2 can be extracted by transmitted illumination. It can also be applied.

  In this embodiment, a blank portion of “license condition etc.” is used as the watermark pattern area 9. However, the present invention is not limited to this, and the range is expanded to the extent that character recognition is not affected. You may do it. In particular, it is possible to detect whether or not the face photo is affixed from above by spreading it over a part of the face photo.

  Furthermore, in the present embodiment, a white LED is used as the second light source 5, but the present invention is not limited to this, and an LED having a hue that can easily enhance the watermark may be used. In the present embodiment, a line sensor is used as the reading unit 6. However, the present invention is not limited to this, and another sensor may be used.

It is a figure which shows the reader of this invention. It is a side view which shows the positional relationship of each light source and a reading means. It is a figure which shows the relationship between a medium and shaving correction data, (A) shows a medium, (B) shows a 2nd shaving correction table, (C) shows a 1st shading correction table. It is a block diagram which shows a control board. It is a time chart which shows operation | movement of a reader. It is a flowchart which shows operation | movement of a reader. It is a figure which shows a gamma correction curve, (A) is a normal gamma correction curve, (B) is an example of the gamma correction curve of the watermark pattern area | region combined when the density of the watermark pattern is distributed on the white side. Show.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Reading apparatus 2 Medium 3 1st light source 5 2nd light source 6 Reading means 9 Watermark pattern area | region 11 1st shaving correction table 12 2nd shading correction table

Claims (4)

A first light source for irradiating information recorded on the surface of the medium from the surface side of the medium, and a watermark reading section for reading a watermark pattern provided on the medium and recorded on the surface of the medium read the information, reading a second light source for morphism irradiation from the back side of the medium in the watermark region is set in a range that does not affect to recognize, the information recorded on the surface of the medium And a reading means for reading the image of the watermark pattern in a state irradiated with light from the second light source, and used for reading information recorded on the surface of the medium outside the watermark reading section. A first shading correction table, and a second shading correction used to read information recorded on the medium surface and the watermark pattern image in the watermark reading section. The second shading correction table is obtained by associating a portion corresponding to the watermark pattern area of the first shading correction table with the watermark pattern, and the watermark reading section is detected. Then, the reading apparatus which can be switched from the first shading correction table to the second shading correction table . A first γ correction curve used when reading information recorded on the surface of the medium outside the watermark reading section, information recorded on the medium surface in the watermark reading section, and an image of the watermark pattern Having a correction table recording a second γ correction curve used to read the image, and switching from the first γ correction curve to the second γ correction curve when the watermark reading section is detected The reading device according to claim 1. 3. The reading apparatus according to claim 1, wherein the reading unit is a line sensor capable of reading image data for at least one line of the medium. The second shading correction table partially increases the amplitude of the correction data in the watermark pattern area range based on the first shading correction table to set the image gain of the watermark pattern area to the range of the watermark pattern area. The reading apparatus according to claim 1, wherein the reading gain is lower than an image gain outside of.

Images