JP4406835B2 - Communication apparatus and communication method - Google Patents

Description

The present invention provides a communication apparatus and communication method, and is preferably related to a communication system that communicates using the unique information.

  Conventionally, a management device that holds unique identification information and a plurality of communication terminal devices that are equipped with a nonvolatile memory that stores the identification information in advance are connected via a predetermined network, and the management device and each communication terminal There has been proposed a communication system in which a device performs predetermined authentication processing using identification information during communication.

In this case, as a technique for preventing theft of identification information, a technique for mounting a non-volatile memory between predetermined members in a deep layer in a communication terminal device, or a non-volatile memory and storing the identification information in the non-volatile memory There is a technique (see, for example, Patent Document 1) in which an encryption / decryption unit that encrypts data only when it is stored as one chip.
JP 2003-256282 A

  By the way, this method not only complicates the installation of the non-volatile memory, but also prevents the identification information from being stolen by the person installing the non-volatile memory. As a result, the reliability of the identification information itself is obtained. Was still inadequate.

The present invention has been made in consideration of the above points, and intends to propose a communication device and a communication method that can improve the reliability of security measures.

The present invention for solving the above problem is a communication device, the output signals output from an imaging element, and removing means for removing the interfering factor for the signal component caused by variation in the structure of the image sensor, Based on the output signal from which the interference factor has been removed by the removing means, the unique information extracting means for extracting the sensitivity pattern unique to the image sensor corresponding to the variation, and authentication between other communication devices using the sensitivity pattern information Authentication means for executing processing . The unique information extracting means detects a luminance deviation state of each pixel in the output signal from which the interference factor is removed, and extracts sensitivity pattern information indicating pixels corresponding to the luminance having a deviation larger than a predetermined reference in a predetermined order. To do.

Further, the present invention is a communication method , wherein a first step of removing an interfering factor with respect to a signal component caused by structural variation of an image sensor from an output signal output from the image sensor , and a first step in on the basis of the output signals interfering factors have been removed, and a second step of extracting the corresponding IMAGING elements inherent sensitivity pattern variation, by using a sensitivity pattern information, with other communication devices And a third step of executing an authentication process . In the second step, the luminance deviation state of each pixel in the output signal from which the interference factor is removed is detected, and sensitivity pattern information indicating pixels corresponding to the luminance having a deviation larger than a predetermined reference is extracted in a predetermined order. To do.

According to the present invention as described above, among the luminance deviation state of each pixel in the output signal of the imaging device interfering factors have been removed for the signal component caused by variation in the structure of the imaging device, than a predetermined reference The authentication processing is executed using information indicating pixels corresponding to luminance having a large deviation in a predetermined order. As a result, it is possible to generate unique information that cannot be known to a third party even at the time of manufacture without being stored in advance in a non-volatile memory or the like. Therefore, it is possible to easily ensure the confidentiality of the unique information. Thus, reliability in security measures can be improved.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

(1) Principle As shown in FIG. 1, a CCD (Charge Coupled Device) imaging device 1 is formed by sequentially laying a substrate 1A having a plurality of photodiodes FD arranged in a grid, an insulating layer 1B, and a light shielding layer 1C. Thus, an opening OP penetrating the insulating layer 1B and the light shielding layer 1C is provided at a position corresponding to the photodiode FD, and a condenser lens CL is provided outside the opening OP. .

  The CCD image pickup device 1 has structural variations represented by the shapes of the opening OP and the condenser lens CL due to various factors in the manufacturing process. In this case, in the CCD imaging device 1, a difference occurs in the shape of the light beam incident on each photodiode FD due to the structural variation, and the amount of incident light in each photodiode FD increases or decreases according to the difference.

  Therefore, the image signal output from the CCD image pickup device 1 includes a high-frequency component (hereinafter referred to as a variation component) due to the structural variation of the CCD image pickup device 1.

  This variation component is treated as unnecessary information (noise) in terms of image quality in the image, but is reflected in a variation state unique to the CCD image sensor 1, in other words, a sensitivity state for each pixel in the CCD image sensor 1. Therefore, it can be handled as useful information from the viewpoint of security measures in a computer network.

However, the image signal output from the CCD image pickup device 1 has a substantial disturbing factor for the variation component (hereinafter referred to as a variation component disturbing factor).
(A) Noise resulting from a difference in brightness in a subject (hereinafter referred to as subject brightness-difference noise)
(B) Noise resulting from a difference in brightness of the optical lens itself (hereinafter referred to as optical lens brightness difference-induced noise)
(C) Noise resulting from a difference in light transmittance for each pixel in a color filter array such as an RGB system or a complementary color system (hereinafter referred to as light transmittance difference-induced noise)
(D) Thermal noise in the CCD image sensor 1 is also included.

  Therefore, the authentication system according to the present embodiment removes the variation component interfering factor from the image signal output from the CCD image sensor 1, and identifies the sensitivity state for each pixel in the CCD image sensor 1 based on the removed image signal. It is characterized in that it is generated as information (hereinafter referred to as sensitivity pattern information).

  Hereinafter, the authentication system according to the present embodiment will be described in detail.

(2) Configuration of Authentication System In FIG. 2, reference numeral 10 denotes an authentication system according to the present embodiment as a whole, and manages a plurality of information terminal devices 11 (11A to 11N) and these information terminal devices 11 (11A to 11N). The management device 12 is connected to each other via a wired or wireless transmission path.

  In this information terminal device 11 (11A to 11N), as a color filter array having a plurality of color pixels that perform color spectroscopy of visible light as units (hereinafter referred to as color spectral pixel units), for example, “R”, “ A CCD image pickup device (hereinafter referred to as a color CCD image pickup device) 13 (13A to 13N) having a color filter array having G, G, and B as color spectral pixel units is mounted.

  The information terminal device 11 (11 </ b> A to 11 </ b> N) generates sensitivity pattern information unique to the color CCD image pickup device 13 (13 </ b> A to 13 </ b> N) during communication with the management device 12, and transmits this sensitivity pattern information to the management device 12. Has been made.

  On the other hand, the management device 12 is connected to an HDD (Hard Disk Drive) 14 in which sensitivity pattern information specific to the color CCD image pickup devices 13 (13A to 13N) mounted on the information terminal devices 11 (11A to 11N) is registered. ing.

  Then, the management device 12 collates the sensitivity pattern information received in the HDD 14 (hereinafter referred to as registered sensitivity pattern information) for each received sensitivity pattern information, and the sensitivity pattern information according to the collation result. The communication process is executed only when it is determined that the color CCD image pickup device 13 (13A to 13N) is mounted on the transmission source that transmitted the.

  As described above, in the authentication system 10, sensitivity pattern information unique to the color CCD image pickup device 13 is used as a validity judgment target in the device that has accessed the management device 12, and the registered sensitivity registered in advance is used. Only the information terminal device 11 (11A to 11N) having the color CCD image pickup device 13 corresponding to the pattern information can communicate with the management device 12.

(3) Configuration of Information Terminal Device Since the information terminal devices 11 (11A to 11N) have the same configuration, only the configuration of the information terminal device 11A will be described here.

(3-1) Basic Configuration As shown in FIG. 3, the information terminal device 11 </ b> A has an operation unit 22, a storage unit 23, and a CCD camera unit 24 with respect to the control unit 21 that controls the entire information terminal device 11 </ b> A. And the communication unit 25 are connected to each other.

  The control unit 21 has a computer configuration including a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), an image processor, and a clock generator, and an imaging command for imaging a subject from the operation unit 22. In this case, the image capturing state of the CCD camera unit 24 is controlled by executing automatic exposure processing and autofocus processing based on the image data D1 output from the CCD camera unit 24 at this time.

  In this state, the control unit 21 converts the image data D1 output from the CCD camera unit 24 into a preset compression format at a predetermined timing, and stores the compressed image data D2 obtained as a result in the storage unit 23. It is made to do.

  As shown in FIG. 4, the CCD camera unit 24 includes a color CCD image pickup device 13A, an optical system 31 that guides incident light L incident on the information terminal device 11A from the outside to the color CCD image pickup device 13A, and The A / D (Analog / Digital) conversion unit 32 converts the image signal S1 output from the color CCD image pickup device 13A into image data D1, and includes the optical system 31, the color CCD image pickup device 13A, and A / D conversion. The image data D1 obtained through the unit 32 sequentially is given to the control unit 21.

  The optical system 31 includes, for example, three lenses 31A, 31B, and 31C and a diaphragm 33 provided between the lenses 31B and 31C, and the diaphragm is operated according to an operation by the user or an automatic exposure process by the control unit 21. In addition to changing the aperture value of 33, the positions of the lenses 31A, 31B, and 31C can be changed in accordance with the zoom operation by the user and the autofocus process by the control unit 21.

  Therefore, the CCD camera unit 24 is capable of adjusting the light amount of the incident light L incident on the color CCD image pickup device 13A and adjusting the focal lengths and focus positions of the lenses 31A to 31C.

  The communication unit 25 (FIG. 3) transmits imaging sensitivity pattern information and other various information given from the control unit 21 to the management device 12 according to a predetermined communication protocol, or transmits information transmitted from the management device 12 to the control unit. 21 is given.

(3-2) Sensitivity Pattern Information Generation Processing In addition to this configuration, the control unit 21 is configured to perform color CCD processing when a communication command for communicating with the management device 12 (FIG. 2) is given from the operation unit 22 or the communication unit 25. Sensitivity pattern information generation processing for generating sensitivity pattern information unique to the image sensor 13A is executed.

Here, when the processing contents of the sensitivity pattern information generation processing by the control unit 21 are functionally classified, as shown in FIG. 5, based on image data D1 (D1 _{1 to} D1 _n ) sequentially output from the CCD camera unit 24. The interference factor removal processing unit 41 that removes the variation component interference factor and the sensitivity pattern information extraction unit 42 that extracts the sensitivity pattern information based on the removal result.

  The interference factor removal processing unit 41 further adds an image pickup control unit 41A for controlling the image pickup state of the CCD camera unit 24 and images sequentially obtained from the CCD camera unit 24 whose image pickup state is controlled by the image pickup control unit 41A. An averaging unit 41B that averages, a high-frequency component extracting unit 41C that extracts a high-frequency component of the image that has been averaged, and a pixel extraction that extracts pixels corresponding to a predetermined color pixel from the extracted high-frequency component image It can be divided into part 41D.

  Hereinafter, imaging control processing by the imaging control unit 41A, addition averaging processing by the addition averaging unit 41B, high frequency component extraction processing by the high frequency component extraction unit 41C, pixel extraction processing by the pixel extraction unit 41D, and sensitivity by the sensitivity pattern information extraction unit 42 The pattern information extraction process will be described in detail.

(3-2-1) Imaging Control Processing The imaging control unit 41A receives incident light L from a monotonous imaging target (hereinafter referred to as a monotonic imaging target) such as a single color or colorless plate or wall. In a state in which (FIG. 4) is incident on the CCD camera unit 24, monotonic imaging is performed based on image data D1 (D1 _{1 to} D1 _n ) sequentially output from the CCD camera unit 24 as an imaging result of the monotonic imaging target. The aperture value of the aperture 33 is varied so that the target shade is smooth as a whole, and the positions of the lenses 31A to 31C are varied.

As a result, the subject contrast noise in the monotonous imaging target among the variation component interference factors is removed from the image data D1 (D1 _{1 to} D1 _n ). Data (hereinafter referred to as low-frequency component image data) D10 (D10 _{1 to} D10 _n ) is sequentially output from the CCD camera unit 24.

Then, the imaging control unit 41A sequentially transmits the low frequency component image data D10 (D10 _{1 to} D10 _n ) to the addition averaging unit 41B at a predetermined timing.

  In this way, the imaging control unit 41A can remove subject light / dark difference noise among the variation component interference factors by controlling the imaging state of the CCD camera unit 24.

(3-2-2) averaging process the averaging section 41B, for example, as shown in FIG. 6, the image BD ₂ based on the image BD ₁ and low-frequency component image data D10 ₂ based on the low-frequency component image data D10 ₁ adds corresponding pixels respectively mean between adds corresponding pixels respectively averaged between the image BD ₃ based on the averaged image and the low-frequency component image data D10 _3.

The averaging section 41B, the remaining low-frequency component image data _D10 4 each image based on the to D10 _{n BD} 4 even with ～BD _n the same manner sequentially added and averaged, the image based on the last low-frequency component image data D10 _n The addition average image AVD between the BDs _n is sent to the high frequency component extraction unit 41C (FIG. 5) as data (hereinafter referred to as addition average image data) D11.

  In this case, the thermal noise in the color CCD image pickup device 13A among the variation component interference factors, as can be seen from FIG. 6, the luminance component becomes lower every time it is sequentially added and averaged. As a result, from the added average image data D11. It will be removed.

In this way, the addition averaging unit 41B can remove thermal noise in the color CCD image pickup device 13A by sequentially averaging the low-frequency component image data D10 (D10 _{1 to} D10 _n ). .

The thermal noise in the color CCD image pickup device 13 has the strongest influence among the variation component disturbing factors included in the image data D1, and therefore the images BD _{1 to} BD _n based on the low-frequency component image data D10 _{1 to} D10 _n . It is desirable to increase the average number of additions as much as possible in consideration of various circumstances such as processing load on the control unit 21.

(3-2-3) High-frequency component extraction processing The high-frequency component extraction unit 41C sequentially calculates the average of pixel groups in units of, for example, 5 pixels in the vertical and horizontal directions in the addition average image data D11 supplied from the addition averaging unit 41B. In this way, LPF (Low Pass Filter) processing is executed.

  Then, the high frequency component extraction unit 41C subtracts the low frequency component of the addition average image data D11 extracted as the LPF processing result from the addition average image data D11, so that the high frequency component in the addition average image data D11 is obtained. Is extracted and sent to the pixel extraction unit 41D as data (hereinafter referred to as high-frequency component data) D12.

  As a result, the high-frequency component data D12 generally removes not only the noise caused by the optical lens contrast difference in the low-frequency component of the data, but also the subject contrast difference noise that could not be removed by the imaging control processing. On the other hand, the variation component of the high frequency component remains.

  In this manner, the high frequency component extraction unit 41C can reliably remove both the subject brightness difference noise and the optical lens brightness difference noise by extracting the high frequency component of the addition average image data D11. Has been made.

(3-3) Pixel Extraction Processing The pixel extraction unit 41D colors pixel data corresponding to, for example, the “G” pixel of the CCD solid-state imaging device 13A among the high frequency component data D12 supplied from the high frequency component extraction unit 41C. Extraction is performed for each spectral pixel unit, and the extracted pixel data group for each color spectral pixel unit is sent to the sensitivity pattern information extraction unit 42 as variation component extraction result data (hereinafter referred to as variation component extraction data) D13. To do.

  As a result, noise due to the difference in light transmittance due to the difference in light transmittance for each color pixel in the color filter array is removed from the variation component extraction data D13.

  In this way, the pixel extraction unit 41D can reliably remove the noise caused by the difference in light transmittance by extracting the pixels corresponding to the predetermined color pixels.

  As described above, the interference factor removal processing unit 41 eliminates the variation component of the image data D1 (subject light / dark difference noise, thermal noise in the color CCD image sensor 13A, optical lens light / dark difference noise and light transmission). Only rate difference-induced noise) can be removed.

(3-4) Sensitivity Pattern Information Extraction Processing The sensitivity pattern information extraction unit 42 sequentially detects the luminance value of the variation component extraction data D13 for each pixel, and based on the detection result, as shown in FIG. A luminance histogram (log scale) representing sensitivity characteristics for each pixel in the CCD image sensor 13A is generated. In FIG. 7, the vertical axis represents the number of pixels, and the horizontal axis represents the standard deviation of luminance values.

  As is apparent from FIG. 7, it can be seen that the deviation is smaller on the plus side than the reference (corresponding to “0” (median value) in FIG. 7), and the deviation is larger on the minus side than the reference. The pixel group having a large deviation is a set of positions having low sensitivity due to structural variations inherent to the color CCD image pickup device 13A, and means a sensitivity pattern corresponding to the variations.

  Then, the sensitivity pattern information extraction unit 42, based on such a luminance histogram, has a low sensitivity pixel (hereinafter referred to as a low sensitivity pixel) that is larger than a preset threshold value (hereinafter referred to as a deviation threshold value). ) Data (hereinafter referred to as low-sensitivity pixel data), and a data string obtained by combining these low-sensitivity pixel data in a predetermined order is generated as sensitivity pattern information D14.

  In this state, the sensitivity pattern information extraction unit 42 performs a predetermined encryption process on the sensitivity pattern information D14, and transmits the encryption sensitivity pattern information obtained as a result to the management apparatus 12 via the communication unit 25. Has been made.

  As described above, the sensitivity pattern information extraction unit 42 is based on the variation component extraction data D13 obtained by removing only the variation component interference factor from the image data D1 output from the manufactured color CCD image pickup device 13A. By generating the sensitivity pattern information D14 corresponding to the structural variation inherent to the CCD image pickup device 13, the sensitivity pattern information D14 is generated in a state that is not known to not only the third party but also the manufacturer of the color CCD image pickup device 13A. Has been made to be able to.

(4) Configuration of Management Device As shown in FIG. 8, the management device 12 controls the control unit 51 that controls the entire management device 12 according to the same communication protocol as the HDD 14 and the communication unit 25 (FIG. 3). The communication unit 52 that transmits and receives information is connected to each other.

  The control unit 51 has a computer configuration including a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), an image processor, and a clock generator, and is based on a control program stored in the ROM. The communication unit 52 and the HDD 14 are appropriately controlled.

  In practice, when the encryption sensitivity pattern information is given via the communication unit 52, the control unit 51 restores the sensitivity pattern information D14 by performing a predetermined decryption process on the encryption sensitivity pattern information, and the restoration The obtained sensitivity pattern information D14 and the registered sensitivity pattern information RF read from the HDD 14 are sequentially collated.

  When the matching result is a matching rate lower than a preset threshold value, the control unit 51 transmits the sensitivity pattern information D14 at this time as the transmission source of the color CCD image pickup device 13 (13A to 13N). Is determined to be not the information terminal device 11 (11A to 11N), and communication with the device is cut off.

  On the other hand, if the coincidence rate is higher than the threshold value, the control unit 51 is an information terminal device in which the transmission source that transmits the sensitivity pattern information D14 at this time is mounted with the color CCD image pickup device 13 (13A to 13N). 11 (11A to 11N), and communication with the information terminal device 11 is started.

  In this way, the control unit 51 executes the authentication process using information (registered sensitivity pattern information RF, sensitivity pattern information D14) reflecting the sensitivity state of each pixel in the color CCD image pickup device 13. .

(4-1) Experimental Results Here, as shown in FIG. 9, a series of sequentially executing the above-described interference factor removal processing and sensitivity pattern information extraction processing for ₁₂₈ images IM _{1 to} IM 128 obtained from the monotonous imaging target. For the RGL (hereinafter referred to as a reference sequence) and n images IM _{1 to} IM _n (n = 1, 2,..., Or 16) obtained from the monotonous imaging target, A sequence (hereinafter referred to as a verification target sequence) CGL for sequentially executing sensitivity pattern information extraction processing is prepared, and registered sensitivity pattern information RF obtained as a result of operating these sequences RGL and CGL under various conditions; The verification result between the sensitivity pattern information D14 was verified.

  FIG. 10 shows the registration threshold value information RF and sensitivity pattern information composed of low-sensitivity pixels that are equal to or higher than the deviation threshold value by appropriately changing the deviation threshold value used as the extraction reference in the sensitivity pattern information extraction process for both the reference sequence RGL and the verification target sequence CGL When D14 is collated, the ratio at which the low-sensitivity pixels (addresses) do not match is shown.

In FIG. 10, the vertical axis represents the mismatch rate (log ₁₀ ), and the horizontal axis represents the standard deviation. In the comparison target series CGL, the interference factor removal process and the sensitivity pattern information extraction process are sequentially executed for the _eight images IM _{1 to} IM ₈ .

  As is apparent from FIG. 10, the discrepancy rate is almost constant regardless of which deviation threshold is selected, and is approximately 10% (10 to the power of −1) to 50% (10 −0. It can be seen that it is between 3rd power). From this, it can be considered that the coincidence rate is between 50% and 90%.

On the other hand, FIG. 11 shows the registration sensitivity pattern information RF and the sensitivity pattern composed of low-sensitivity pixels having a certain deviation threshold value or more by appropriately changing the number of addition processes of the images IM _{1 to} IM _n in the addition averaging process of the comparison target series CGL. When the information D14 is collated, the ratio at which the low-sensitivity pixels do not match is shown. In FIG. 11, the vertical axis represents the mismatch rate (log ₁₀ ), and the horizontal axis represents the addition average number.

  As is apparent from FIG. 11, the mismatch rate when the average number of additions is 1 is about 50% (10 to the power of −0.3), whereas the mismatch rate when the average number of additions is 16 times. It can be seen that the rate has dropped to about 15% (10 to the power of −0.8). From this, it can be considered that a matching rate of 50% can be secured at worst.

(4-2) Consideration Next, on the basis of the verification result, the probability of misjudging it as another CCD image pickup device even though it is the same CCD image pickup device (hereinafter referred to as a legitimacy identification failure probability). Then, the probability of misjudging the same CCD image sensor even though it is a completely different CCD image sensor (hereinafter, referred to as an unfair discrimination failure probability) will be considered.

(4-2-1) Validity Identification Failure Probability First, as the validity identification failure probability, the verification result when the sensitivity pattern information D14 and the registered sensitivity pattern information RF obtained from the same CCD image sensor are compared is predetermined. The probability of being less than the threshold is obtained.

  That is, such an event occurs when the number of matching low-sensitivity pixels is p, the number of low-sensitivity pixels that do not match is f, the probability that one low-sensitivity pixel matches is pp, and the probability that it does not match is pf. The probability P of

The number N when such an event occurs is a combination ( ₁₀₀₀ C _p ) from which ₁₀₀₀ is selected, and the following formula:

Can be expressed as

  Therefore, the correctness identification failure probability is expressed by the following equation that summarizes the equations (1) and (2).

Can be obtained.

Here, as described above with reference to FIG. 11, it can be considered that a matching rate of 50% can be secured at worst, so that “pp” = pf = 0.5 and “1000” low-sensitivity pixels whose deviations are greatly different from each other. It is assumed that if the coincidence ratio of the low-sensitivity pixels is less than 10% (100/1000) when the registered sensitivity pattern information RF and the sensitivity pattern information D14 consisting of When the expression (3) is calculated, the correctness identification failure probability becomes approximately 0.7 ^* 10 ⁻¹⁶¹ (a state in which a wrong judgment is made only 7 times with 10 ⁻¹⁶¹ times of the matching result).

(4-2-2) Inappropriate discrimination failure probability Next, as the inadequate discrimination failure probability, the verification result when the sensitivity pattern information D14 and the registered sensitivity pattern information RF obtained from different CCD image sensors are compared with each other is obtained. A probability of being equal to or higher than a predetermined threshold is obtained.

  That is, when the number of pixels of an image is m, the number of low-sensitivity pixels selected from the image is n, and the number of matched low-sensitivity pixels is p, the number of low-sensitivity pixels that do not match is (m− A combination (m-pCn-p) in which (p) to (np) are selected.

  In addition, the number in the case where the number of pixels that coincides with the number of pixels m is a combination (nCp) in which p is selected from n. Therefore, since the number of p elements that match is the product of both, (m−nCn−p * nCp).

  On the other hand, since the number in all cases is (mCn), the probability P of such an event is given by

The probability of unfair identification failure when P or more matches is given by

Can be obtained.

  Here, the coincidence rate of the low sensitivity pixels when the registered sensitivity pattern information RF and the sensitivity pattern information D14 composed of “1000” low sensitivity pixels having greatly different deviations are compared with each other is less than 10% (100/1000). Assuming that it is determined as another CCD imaging device in the case of m, m is 10000000, n is 1000, and p is 100. Substituting these into equation (5) gives

It becomes. When this equation (6) is calculated, the unfair identification failure probability becomes approximately 0.4 ^* 10 ⁻¹⁶⁴ (a state in which erroneous determination is made only 4 times with 10 ⁻¹⁶⁴ verification results).

  Strictly speaking, it is necessary to integrate p up to 1000. However, since p has a relationship of decreasing the number of digits by 2 when it is increased by 1, there is no necessity for practical use.

  As can be seen from the above consideration results, the probability of misjudgment is extremely small, so that sensitivity pattern information (RF, D14) can be treated as an identification target without any practical problem. Therefore, it can be said that the information terminal device 11 can generate highly reliable sensitivity pattern information (RF, D14) as an identification target.

(5) Operation and effect according to the present embodiment In the above configuration, the information terminal device 11 in the authentication system 10 includes the structure of the color CCD image pickup device 13 among the image data D1 output from the color CCD image pickup device 13. The variation component interfering factor for the variation component due to the above variation is removed.

  Then, the information terminal device 11 extracts the sensitivity state in the color CCD image sensor 13 corresponding to the structural variation as sensitivity pattern information D14 based on the variation component extraction data D13 obtained as a result of this removal.

  Therefore, the information terminal device 11 can generate the sensitivity pattern information D14 that cannot be known to a third party even during production without being stored in advance in a nonvolatile memory or the like. Even when the removal method and the sensitivity state extraction method in the color CCD image pickup device 13 are disclosed to a third party, the confidentiality of the sensitivity pattern information D14 itself can be easily ensured.

  In addition to this, the information terminal device 11 can prove the correspondence between the sensitivity pattern information D14 and the color CCD image pickup device 13, and can easily detect the presence or absence of a counterfeit product.

  According to the above configuration, the variation component interfering factor is removed from the image data D1 output from the color CCD image pickup device 13, and the sensitivity pattern corresponding to the structural variation of the color CCD image pickup device 13 based on the removal result. By extracting the information D14, it is possible not only to easily secure the confidentiality of the sensitivity pattern information D14, but also to easily detect the presence or absence of counterfeit products, and thus the reliability in security measures Can be improved.

(6) Other Embodiments In the above-described embodiment, as a removing means for removing an interfering factor for a signal component caused by structural variation of the image sensor from the output signal output from the image sensor. Although the subject light-dark difference noise, optical lens light-difference noise, light transmittance difference-induced noise, and thermal noise interference factors (variation component interference factors) in the color CCD image pickup device 13 have been described. The present invention is not limited to this, but in addition to these interference factors, noise (hereinafter referred to as image sensor drive noise) or A / D conversion that occurs when the color CCD image sensor 13 is driven based on the clock signal. Noise generated in the device 32 itself (hereinafter referred to as A / D conversion noise) may be removed.

  In this case, the image sensor driving noise and the A / D conversion noise have a certain reproducibility. Therefore, conflicting data of the measurement results of the image sensor driving noise and A / D conversion noise can be stored in advance in the memory in the control unit 21. Therefore, the control unit 21 subtracts the image sensor driving noise and the A / D conversion noise by subtracting the data from any one of the imaging control processing result, the addition averaging processing result, the high frequency component extraction processing result, or the pixel extraction processing result. Can be removed.

  If the image sensor driving noise and the A / D conversion noise are removed in addition to the variation component interference factor in this way, the extraction accuracy of the sensitivity pattern information (RF, D14) can be improved. Reliability in security measures can be improved.

  Further, as the removing means, interference factors other than these image sensor driving noise and A / D conversion noise may be removed, and for example, it may be replaced with light transmittance difference caused noise. In short, if the thermal noise in the color CCD image pickup device 13 having the strongest influence is removed, other interference factors can be arbitrarily selected. Even in this case, the above-described embodiment is used. The same effect can be obtained.

  Further, in the above-described embodiment, the interference factor removal processing unit 41 performs the variation as the specific information extracting unit that extracts the sensitivity state of the image sensor corresponding to the variation as the specific information based on the output signal removed by the removing unit. A luminance histogram is generated with the vertical axis representing the number of pixels and the horizontal axis representing the standard deviation of the luminance value for the variation component extraction data D13 obtained as a result of the removal of the interfering factor. Although the case where the pixel corresponding to the luminance with a large deviation is extracted as the sensitivity pattern information D14 has been described, the present invention is not limited to this, and the luminance deviation state in the variation component extraction data D13 is not limited to this. And detecting the sensitivity state of the image sensor corresponding to the variation based on the detection result. Kill.

Furthermore, in the above-described embodiment, as an integration unit that integrates a plurality of output signals sequentially output from the image sensor as the imaging result of the monotonic imaging target so that the density of the monotonic imaging target is smooth as a whole, Although the case where a plurality of image data D10 (D10 _{1 to} D10 _n ) sequentially output from the color CCD image pickup device 13 as the imaging result of the monotonous imaging target is averaged is described, the present invention is not limited to this. For example, three-dimensional DNR (digital noise reduction) may be performed, and portions of the image data D10 (D10 _{1 to} D10 _n ) having a predetermined luminance level or less may be combined. by integrating a plurality of image data _D10 to (D10 1 to D10 _n) as monotonous imaging target shading becomes smooth as a whole There.

Further, in the above-described embodiment, the image data D1 (D1 ₁ to D1 ₁ to D1) sequentially output from the CCD camera unit 24 in a state where the incident light L (FIG. 4) from the monotonic imaging target is incident on the CCD camera unit 24. D1 _n ), the automatic exposure process and the autofocus process are executed to make the gradation of the monotonic imaging target smooth as a whole. However, the present invention is not limited to this, and the monotonic imaging target is not limited to this. By providing a scattering plate or the like that uniformly scatters the incident light L in front of the lens 31A, the density of the monotonic imaging target may be smooth as a whole.

  Further, in the above-described embodiment, the case where the information terminal device 11 (color CCD image sensor 13) is identified using only the sensitivity pattern information (RF, D14) has been described, but the present invention is not limited to this. Instead, the information terminal device 11 (color CCD image sensor 13) may be identified using a combination of the sensitivity pattern information (RF, D14) and other identification information.

  Specifically, for example, the manufacturing number information of the information terminal device 11 is registered in advance in the HDD 14 of the management device 12 in association with the registration sensitivity pattern information RF of the color CCD image pickup device 13. Then, the information terminal device 11 transmits the sensitivity pattern information D14 and the manufacturing number information to the management device 12 during communication, and the management device 12 collates both the sensitivity pattern information D14 and the manufacturing number information with the registered contents of the HDD 14.

  In this way, since the management device 12 can prove the correspondence between the color CCD image pickup device 13 and the information terminal device 11, for example, the color CCD image pickup device 13 itself is stolen and the information terminal device 11 is stolen. It is possible to prevent impersonation due to being mounted on a device other than the above, and as a result, it is possible to more easily detect that the product is a counterfeit product.

  In addition to using the sensitivity pattern information D14 as an identification target, the sensitivity pattern information D14 may be used as an element on the encryption system such as an encryption key or a random number source (seed). Even in this way, reliability in terms of security measures can be improved as in the above-described embodiment.

  Furthermore, in the above-described embodiment, a case has been described in which the control unit 21 performs the above-described imaging control processing, addition averaging processing, high-frequency component extraction processing, pixel extraction processing, and sensitivity pattern information extraction processing by software. However, the present invention is not limited to this, and part or all of these processes may be executed by dedicated hardware, and the above-described verification process may be incorporated depending on the application of security measures on the network. Also good.

  INDUSTRIAL APPLICABILITY The present invention can be used as a security measure in a network, for example, when a terminal device such as a personal computer or a mobile phone, or a device such as a home electronic device is distinguished from an external device.

It is a basic diagram which shows the structure of a CCD image pick-up element. It is block s which shows the structure of the authentication system by this Embodiment. It is a block diagram which shows the structure of an information terminal device. It is a basic diagram which shows the structure of a CCD camera part. It is a functional block diagram which shows the sensitivity pattern information generation process by a control part. It is an approximate line figure used for explanation of addition average processing. It is a basic diagram which shows the sensitivity characteristic for every pixel in a color CCD image sensor. It is a block diagram which shows the structure of a management apparatus. It is a basic diagram with which it uses for description of collation experiment. It is a basic diagram which shows the relationship between a deviation threshold value and the mismatch rate of a collation result. It is a basic diagram which shows the relationship between an addition average number and the mismatch rate of a collation result.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Authentication system, 11A-11N ... Information terminal device, 12 ... Management device, 13A-13N ... Color CCD image sensor, 14 ... HDD, 21, 51 ... Control part, 22 ... Operation part, 23 …… Storage unit, 24 …… CCD camera unit, 25, 53 …… Communication unit, 31 …… Optical system, 32 …… A / D conversion unit, 33 …… Aperture, 41 …… Interference factor removal processing unit 41A: Imaging control unit, 41B: Addition averaging unit, 41C: High frequency component extraction unit, 41D: Pixel extraction unit, 42: Sensitivity pattern information extraction unit.

Claims (2)

Of the output signal outputted from an imaging element, and removing means for removing the interfering factor for the signal component caused by variation of the structure of the image pickup device,
Specific information extraction means for extracting sensitivity pattern information specific to the image sensor corresponding to the variation based on the output signal from which the interference factor has been removed by the removal means ;
Using the sensitivity pattern information , comprising authentication means for executing authentication processing with other communication devices ,
The specific information extracting means is
Extracting the sensitivity pattern information indicative of detecting the brightness deviation state of each pixel in the upper Symbol interferent output signal has been removed, the pixels corresponding to the larger luminance deviation than a reference of Jo Tokoro in a predetermined order Communication device. A first step of removing an interfering factor for a signal component caused by structural variation of the image sensor out of an output signal output from the image sensor;
Based on the output signals interfering factors in the first step has been removed, a second step of extracting the corresponding IMAGING elements inherent sensitivity pattern information to the variation,
Using the sensitivity pattern information, a third step of executing an authentication process with another communication device ,
In the second step,
Extracting the sensitivity pattern information indicating detecting the brightness deviation state of each pixel in the upper Symbol interferent output signal has been removed, the pixels corresponding to the larger luminance deviation than a reference of Jo Tokoro in a predetermined order Communication method.

Images