JPH10285595A - Image data transmitter-receiver - Google Patents

Abstract

PROBLEM TO BE SOLVED: To simplify a processing on a reception side and to perform highly accurate image recognition in the case of compression-transmitting and recognizing images. SOLUTION: In this image data transmitter 1, the images for constituting moving images are discrete cosine transformed (discrete sine transformed) in a first transformation part 15a, the images for constituting still images are discrete sine transformed (discrete cosine transformed) in a second transformation part 15b and they are transmitted. In this image data receiver 2, discrete Fourier data are prepared from the images for constituting the still images to which transformation is executed and the images for constituting the moving images to which the transformation is executed which are pre-images for the images for constituting the still images in a discrete Fourier data preparation means 26 and the images for constituting the still images and the images registered beforehand are compared and judged by using the discrete Fourier data in an image matching means 27.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image data transmitting / receiving apparatus for compressing and transmitting an image and recognizing the image, or a computer-readable recording medium storing a program for realizing the apparatus by a computer.

[0002]

2. Description of the Related Art In recent years, systems and devices utilizing image transmission have been actively developed. For example, such systems as a video conference system in which an image is added to a conventional telephone and a moving image or a quasi-moving image is used, or the Internet A communication system that transmits video together with character information by using a communication method. With the development of such devices and systems,
H261 and JPE defined by the International Telecommunication Union (ITU)
Standardization and standardization of image compression and transmission technologies such as G and MPEG are progressing, and standardization of processing technologies with a higher compression ratio is also being performed. On the other hand, the technology for processing and analyzing the image itself and examining the content of the image has further advanced,
For example, devices and systems that detect only characters from an image containing character information and determine the contents thereof, and that match an image pattern registered in advance to recognize what kind of image is being developed, have been dramatically advanced.

[0003] From such a trend, it is conceivable in the future to develop a technique for simultaneously performing image compression transmission and image recognition. For example, JP-A-6-231254 discloses JPE
By using the technology standardized by G, etc., the compressed and transmitted image is divided into blocks, the average value is obtained, the average value data group is set as one block, and the spatial frequency distribution of the data is examined. A technique for recognizing and determining whether an image is a proper image has been disclosed.

[0004]

According to this technique, compression transmission and recognition of an image can be performed at the same time. However, image recognition is performed by matching an approximate image obtained by blocking an image and taking an average value thereof. Accurate image recognition is not possible. In addition, since the image is divided into blocks by image transmission and data is transmitted after discrete cosine transformation, the receiving side can easily obtain the average value of the block from the discrete cosine transformed data. , It is necessary to check the spatial frequency distribution, so that the processing time is correspondingly increased.

The present invention has been made in view of the above problems, and has been made to enable accurate image recognition when performing both image compression transmission and image recognition, and to simplify processing for examining a spatial frequency distribution on a receiving side. Aim.

[0006]

In order to solve the above-mentioned problems, the present invention comprises an image data transmitting device and an image data receiving device, wherein the image data transmitting device obtains a temporally continuous image. An acquisition unit, and an image forming a moving image or a still image based on a magnitude of a change amount between one of the obtained images and a previous image of the target image. Image type determining means for determining whether or not the image type. A first conversion unit that performs a discrete cosine transform or a discrete sine transform on the target image determined to be an image forming a moving image; and a first conversion unit that determines a target image determined to be a still image. A second cosine transformation unit that performs a different transformation from the first cosine transformation unit, the second cosine transformation unit performing a transformation different from that performed by the first transformation unit, and transmits the target image transformed by the first or second transformation unit; Means.

[0007] When the target image converted and transmitted by the first conversion means is an image constituting a moving image, the target image includes image data of only a difference from the previous image. Further, the image data receiving apparatus according to the present invention is an image that is temporally continuous, and a discrete cosine transform or a discrete sine transform is performed on an image forming a moving image,
Of the discrete cosine transform and the discrete sine transform for images constituting still images, it is also assumed that an image transmitted after being subjected to a transform different from the transform applied to the image constituting the moving image is received. And receiving means for receiving image data, an image constituting the transformed still image, and an image constituting the transformed moving image which is a preceding image of the transformed image. , And image matching means using the discrete Fourier data to compare and determine the spatial frequency of the image forming the still image and the previously registered image.

Here, the discrete Fourier data includes:
Not only the discrete Fourier data itself but also information including a combination of coefficients of a real part and an imaginary part of each term of the discrete Fourier data is used. Furthermore, an image data transmitting / receiving system according to the present invention is a combination of the image data transmitting device and the image data receiving device.

A computer-readable recording medium according to the present invention stores a program for realizing the image data transmitting device or the image data receiving device by a computer.

[0010]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a functional block diagram of the image data transmission / reception system according to the present embodiment. This image data transmitting / receiving system includes an image data transmitting device 1 and an image data receiving device 2.

The image data transmitting apparatus 1 includes an image acquiring unit 1
1, first image memory 12, second image memory 13, image type determination unit 14, first conversion unit 15a, second conversion unit 15b,
It comprises a quantization unit 16, a Huffman coding unit 17, and a transmission unit 18. The image obtaining unit 11 is a part that obtains temporally continuous images. Here, a composite color moving image that is not Y / C separated and captured by a camera (not shown) and the moving image are temporarily stopped. Get a still image. Of course, instead of processing a color image as a composite signal, an image separated by Y / C (a luminance signal and two color difference signals) may be used. The number of pixels to be sampled here is 768 horizontal × 512 vertical pixels.
Needless to say, this is not a limitation. In addition,
It is assumed that the operation of temporarily stopping and canceling the moving image can be performed on the image data receiving device 2 side. Further, the image acquisition unit 11 converts the acquired image data into 8-bit digital data so that the image data can be digitally processed.

The first image memory 12 is a memory for storing image data acquired by the image acquisition section 11 with a horizontal pixel number of 768 × 512 pixels and a quantization of 8 bits.
The image memory 13 is a memory having 768 pixels in the horizontal direction, 512 pixels in the vertical direction, and 8 bits for quantization, which stores the target image recorded in the first image memory 12 in comparison with the previous image and stores only the change. .

The image type determination unit 14 determines whether the target image is an image constituting a moving image or a still image, based on the amount of change between the target image and the previous image. Specifically, pixel information P (x, y) corresponding to a change between the target image and the previous image recorded in the second image memory 13
[X = 1-768, y = 1-512], and the following equation
Using (1), if the following equation is satisfied, the target image is determined to be an image forming a moving image, and if not, the target image is determined to be an image forming a still image. In the following equation (1), T
Is a preset threshold, which is an empirical value determined according to the mode of an image acquired by the image acquisition unit 11. Also, P
(X, y) is the amount of change and is always a positive sign.

[0014]

(Equation 1) The first conversion unit 15a performs a discrete cosine transform on a target image determined to be an image forming a moving image. Specifically, the first conversion unit 15a converts the data from the second image memory into 16 ×
16 pixels are defined as one block, and in block units,
A discrete cosine transform represented by the following equation is performed.

[0015]

(Equation 2) The second conversion unit 15b performs a discrete sine transform on the target image determined to be a still image, and specifically converts the data from the first image memory into a 16 ×
16 pixels are defined as one block, and in block units,
The discrete sine transform represented by the following equation is performed.

[0016]

(Equation 3) The quantization unit 16 divides each coefficient of the image data subjected to the discrete cosine transform by the first transform unit 15a or the image data subjected to the discrete sine transform by the second transform unit 15b by a predetermined divisor to round the remainder. In this way, quantization is performed. Further, the Huffman encoding unit 17 performs so-called Huffman encoding, which receives the data from the quantization unit 16 and allocates codewords having a short code length in order of data having a statistically high frequency of occurrence to reduce the amount of information. The transmission unit 18
Is "0" if the image from the Huffman encoding unit 17 is determined to be an image constituting a moving image by the image type determination unit 14;
1 ′ is transmitted to the image data receiving device 2.

Next, the configuration of the image data receiving device 2 will be described. The image data receiving device 2 includes a receiving unit 21, a decoding unit 22, an inverse quantization unit 23, a received image memory 24, a moving image reproducing unit 25, a discrete Fourier data creating unit 26, and an image matching unit 27. The receiving unit 21 receives the data transmitted from the transmitting unit 18 of the image data transmitting device 1, and the image forming the moving image or the image configuring the still image added by the transmitting unit 18 of the image data transmitting device 1. The information represented by "0" or "1" indicating the information is separated. Further, the decoding unit 22 converts the data received by the receiving unit 21 into the Huffman encoding unit 1 of the image data transmitting apparatus 1.
7 is performed in reverse.

The inverse quantization unit 23 receives the data from the decoding unit 22 and performs the inverse processing of the quantization unit 16 of the image data transmitting device 1. In addition, when the information separated by the reception unit 21 is “0”, that is, when the target image forms a moving image, the inverse quantization unit 23 transmits data to the reception image memory 24 and
1 is “1”, that is, when the target image forms a still image, the discrete Fourier data creation unit 26
Transmit data to

The received image memory 24 adds and stores the data received from the inverse quantization unit 23. That is, as for the images constituting the moving image, only the change from the previous image is transmitted, and by adding this, the restored original image is stored. The received image memory 24 has the same capacity as the first image memory 12. Video playback unit 25
Performs an inverse discrete cosine transform represented by the following equation (4) while reading data from the received image memory 24,
An image is reproduced on a monitor (not shown) using the converted data.

[0020]

(Equation 4) In the moving image reproducing unit 25, since the temporal change of the image is transmitted, the data is added and written to the received image memory 24, the data is read out, and the image is reproduced. There is no problem even if the same processing is performed by providing the third image memory.

The discrete Fourier data creating section 26 creates discrete Fourier data from the transformed image constituting the still image and the preceding image constituting the transformed moving image. . Specifically, the discrete Fourier data creation unit 26 firstly outputs the image data constituting the still image from the inverse quantization unit 23, that is, the image data transmission device 1
Receives substantially the same data as the data output after being subjected to discrete sine transform by the second transform unit 15b, and an image constituting a moving image that has been subjected to discrete cosine transform by the first transform unit 15a recorded in the received image memory 24. Receive almost the same image data.

The image data constituting the moving image subjected to the discrete cosine transform recorded in the received image memory 24 is the image data located at the boundary from the moving image to the still image, and is received by the discrete Fourier data creating unit 26. It can be considered as the same image as the still image. That is, the discrete Fourier data creation unit 26 receives both the data obtained by performing the discrete sine transform on the same image and the data obtained by performing the discrete cosine transform on the same image. The data subjected to the discrete cosine transform and the data subjected to the discrete sine transform correspond to the coefficients of the imaginary part and the real part of the discrete Fourier data when the discrete Fourier transform is performed, respectively.

The image matching unit 27 uses the discrete Fourier data to compare and determine the spatial frequency of the target image and the previously registered image. The spatial frequency comparison between images can be determined using only discrete cosine transform data or only discrete sine transform data.However, accuracy is one-handed, and discrete Fourier data composed of cosine and sine components is used. Analyzing and comparing the spatial frequency of an image can perform a highly accurate comparison and determination.

More specifically, the discrete Fourier data creation unit 2
Using the discrete Fourier data created in step 6, D (x, y) represented by the following equation is created for each block. This D (x, y)
Indicates the length of the vector in Fourier space.

[0025]

(Equation 5) Note that discrete Fourier data is also created for the image to be compared, D (u, v) represented by the above equation is created, and recorded in the image matching unit 27 in advance. Subsequently, whether or not the images match is compared and checked for each block using the following equation.

[0026]

(Equation 6) Where Di (u, v) is D (u, v) for the target image, and
(u, v) is D (u, v) related to the recorded image to be compared.
v). The value K to be compared is an empirical value,
Here, 0.8 is used.

If the above expression is satisfied, it is determined that the image determined to be compared with the target image matches in the block. If the above expression is not satisfied, it is determined that the image determined to be compared with the target image does not match in the block. Is determined. Then, if the number of matching blocks is equal to or more than a predetermined ratio, it is determined that the image to be compared with the target image matches, and if it is less than the predetermined ratio, it is determined that they do not match. Here, it is assumed that the determination is made based on whether or not 80% of the blocks match as a predetermined ratio.
However, this value is an empirical value, and varies depending on the situation.

Here, the spatial frequencies are compared using the length of the vector in the Fourier space, but the spatial frequencies are compared by various methods such as comparing by adding a phase comparison. It is possible. The image data transmitting device and the image data receiving device having the above configurations can be realized by incorporating a program for realizing the functions described above in a general-purpose computer, and the program can be read by a computer. It can be recorded on a recording medium such as a compact disk.

Next, the operation of the image data transmitting / receiving system having the above configuration will be described. First, an operation related to image data compression transmission by the image data transmission device 1 will be described with reference to the drawings. FIG. 2 is a flowchart showing the operation of the image data transmitting apparatus. First, an image captured by a camera (not shown) is sampled by the image acquisition unit 11, taken in for one screen, and stored in the first image memory as 8-bit data (S101). Then, the sampled image is compared with the previous image to create a difference image and stored in the second image memory (S102).

Next, the image type determination unit 14
The addition processing is performed on the image data P (x, y) in the image memory, and the above-mentioned equation (1) is used by using a preset threshold value.
It is determined whether the sampled target image is an image forming a moving image or a still image (S103).
If the target image is determined to be an image constituting a moving image (flag '0'), the first conversion unit 15a converts the image data stored in the second image memory into 16 × 16 pixels as one block, The discrete cosine transform is performed using the unit (2) (S104, S105).

On the other hand. When it is determined that the target image is an image constituting a still image (flag '1'), the second conversion unit 15b converts the image data stored in the first image memory into 16 × 16 pixels by one block. And performs the discrete sine transform shown in Expression (3) in block units (S104,
S106). Here, the image constituting the still image appears when the moving image is temporarily stopped on the receiving side.

If it is determined by the above processing that the target image is an image constituting a moving image, the differential image data is subjected to discrete cosine transform, and normal data compression transmission can be performed. If it is determined that the image is a still image, the entire image data can be subjected to discrete sine transform and transmitted, and used for image recognition.

Subsequently, the quantization unit 16 quantizes the image data that has been subjected to the discrete cosine transform or the discrete sine transform (S107), and the Huffman encoding unit 17 performs the quantization.
Is subjected to Huffman coding (S10).
8). Lastly, the transmission unit 18 adds information “0” to an image constituting a moving image and information “1” to an image constituting a still image in the Huffman-coded image data, and transmits the image data to the image data receiving device 2. I do.

Next, the operation of the image data receiving apparatus 2 relating to reception, decompression or image recognition of image data will be described. FIG. 3 is a flowchart showing the operation of the image data receiving device 2. First, the receiving unit 21 receives data from the image data transmitting apparatus 1, and separates information “0” or “1” indicating the image type (S201). then,
The decoding unit 22 performs inverse processing of Huffman coding on only the image data to decode the data (S202). Further, the data decoded by the inverse quantization unit 23 is subjected to a process opposite to the quantization process performed by the quantization unit 16 of the image data transmitting apparatus 1 (S203).

The inverse quantization unit 23 determines the information separated by the receiving unit 21 and transmits data “0”, that is, if the image constitutes a moving image, the received image memory 24 transmits the data. The addition is stored (S205). Further, the moving image reproducing unit 25 performs an inverse discrete cosine transform represented by Expression (4) while reading the data from the received image memory 24, and reproduces an image (S206).

On the other hand, if the information separated by the receiving unit 21 is “1”, that is, if the image forms a still image, the inverse quantization unit 23 transmits the data to the discrete Fourier data creating unit 26, and The received discrete Fourier data creation unit 26 creates discrete Fourier data together with the data of the previous image recorded in the received image memory (S207). Then, the image matching unit 27 calculates an expression from the discrete Fourier data
D (x, y) is created for each block from (5), the spatial frequency is compared and determined for each block with the reference image registered in advance by Equation (6), and the target image and the reference image are determined based on the proportion of matching blocks. Is determined (S208).

In the present invention, as shown in the above-described operation, an image is divided into a moving image state and a still image state, and in the moving image state, sampling / quantization and discrete cosine transformation are performed by discrete cosine transformation in the same manner as JPEG or MPEG. , Recompression, and Huffman coding, to perform data compression and transmission, and to perform the inverse of these processes to perform data decompression and image reproduction. In a still image state, the image data is subjected to discrete sine transform and transmitted, and discrete Fourier data is created together with the discrete cosine transformed image data stored before the data reception, and collated with the reference image.

In such an operation, it is true that the image cannot be collated in the moving image state. However, when actually recognizing an image such as a camera image, the image is set to a still image state in order to eliminate the influence of an afterimage or the like. In many cases, the image is deliberately stopped and the judgment is made by the machine after confirming with the eyes, which is practical. Further, according to the above-described operation, a moving image is reproduced, but an image determined to constitute a still image is not reproduced. However, if the last image of the moving image is displayed as it is even if it becomes a still image, there is no problem as an image for viewing.

As a specific example of the use of the image data transmission / reception system, for example, an image captured by a camera of a product flowing in a line of a factory or the like is captured by the image data transmission device 1 and displayed by the image data reception device 2. Then, the monitoring side monitors the product for abnormalities on the receiving side. In this case, the video relating to the product flowing through the line is recorded in advance in the image data receiving device 2 as a reference image, and when the observer finds a product that seems to be abnormal while watching the moving image, the image should be stopped. Can be used to determine whether or not the image matches the reference image. Also, by recording a plurality of reference images, it is possible to search for a reference image that matches when monitoring a plurality of targets.

In the present embodiment, as in the case of the MPEG format, for a moving image, only the temporal change of the image is transmitted in the image transmission processing, and the receiving side performs addition processing to the previous image data. Alternatively, a method may be used in which a moving image is transmitted one image at a time and the receiving side does not perform addition processing, such as image transmission in the JPEG format. Further, in the present embodiment, the image is divided into a plurality of blocks and compression transmission and image comparison are performed. However, the size of one block may be made large, and compression and transmission and image comparison may be performed.

Further, in this embodiment, the discrete cosine transform is applied to the image constituting the moving image, and the discrete sine transform is applied to the image constituting the still image. Alternatively, discrete sine transform is performed to form a still image, but discrete cosine transform may be performed on an image. In the present embodiment, the timing of image recognition is such that the image is always recognized when the image is in a still image state. However, the image recognition is performed only when an internal or external trigger is appropriately applied to the image matching means. It can also be recognized.

[0042]

As described above, the present invention has the following effects. First, in the image data transmitting apparatus according to the present invention, a temporally continuous image is acquired by the image acquiring unit, and one of the images acquired by the image type determining unit is set as a target image, and the target image and the image before the target image are acquired. From the magnitude of the change from the image, it is determined whether the target image is an image forming a moving image or an image forming a still image. And performing a discrete cosine transform or a discrete sine transform on the target image determined to be an image constituting the moving image by the first conversion unit,
In addition, of the discrete cosine transform and the discrete sine transform, a transform different from the transform performed by the first transform unit is performed on the target image determined to be a still image by the second transform unit. Then, the target image converted by the first or second conversion unit is transmitted by the transmission unit.

On the other hand, in the image data receiving apparatus according to the present invention, the receiving means is an image which is continuous in time, and an image constituting a moving image is subjected to discrete cosine transform or discrete sine transform to obtain a still image. Of the discrete cosine transform and the discrete sine transform of the image forming the moving image, the image transmitted after the conversion different from the conversion performed on the image forming the moving image is received. And an image constituting the still image to which the discrete Fourier data creating means has performed the conversion,
Discrete Fourier data is created from the image constituting the moving image which has been subjected to the transformation, which is the previous image of the image, and the image matching means registers the image constituting the still image in advance using the discrete Fourier data. The spatial frequency of the obtained image is compared and determined.

By combining the image data transmitting device and the image data receiving device performing the above-described operations, first, the image data transmitting device forms image data constituting a moving image and a still image temporally continuous with the image data. The image data is transmitted after being subjected to one of the discrete cosine transform and the discrete sine transform so that the transform is different from each other. And
Discrete Fourier data that is generated by the image data receiving device and that forms the temporally adjacent moving image and the still image that have been subjected to the conversion, ie, the same image data that has been subjected to two conversions. Can be compared with the reference image with high accuracy.

Further, in the image data transmitting apparatus, the image can be subjected to the compression transmission by performing the discrete cosine transform or the discrete sine transform, and the image data receiving apparatus transmits the image to the still image. Since it is possible to create discrete Fourier data using the incoming data as it is, it is possible to perform spatial frequency analysis without newly performing a discrete Fourier transform. Image analysis and recognition.

As described above, the image data transmitting / receiving apparatus according to the present invention transmits a number of video images through one line,
It can be used for a system that judges images as needed, a system that connects a video terminal to an image database system, and searches for images themselves, and a surveillance system that uses the Internet. is there.

[Brief description of the drawings]

FIG. 1 is a diagram showing a functional block diagram of an image transmission / reception system according to an embodiment of the present invention.

FIG. 2 is a flowchart showing an operation of the image data transmitting apparatus according to the embodiment of the present invention.

FIG. 3 is a flowchart illustrating an operation of the image data receiving apparatus according to the embodiment of the present invention.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 image data transmission device 2 image data reception device 11 image acquisition unit 12 first image memory 13 second image memory 14 image type determination unit 15a first conversion unit 15b second conversion unit 16 quantization unit 17 Huffman encoding unit 18 transmission Unit 21 reception unit 22 decoding unit 23 inverse quantization unit 24 received image memory 25 moving image reproduction unit 26 discrete Fourier data creation unit 27 image matching unit

Claims (5)

[Claims] An image acquisition unit for acquiring an image that is temporally continuous; and an image acquisition unit that acquires one of the acquired images as an object image from an amount of change between the object image and a previous image of the object image. Image type determining means for determining whether the image is a moving image or a still image, and a discrete cosine transform or a discrete sine transform on the target image determined to be a moving image. And a second cosine transform that performs a transform of the discrete cosine transform or the discrete sine transform on the target image determined to be an image forming the still image, the transform being different from the transform performed by the first transform unit. An image data transmission device, comprising: a conversion unit; and a transmission unit that transmits the target image converted by the first or second conversion unit. 2. A discrete cosine transform or a discrete sine transform is applied to an image constituting a moving image, which is a temporally continuous image, and a discrete cosine transform and a discrete cosine transform are applied to an image constituting a still image. An image data receiving apparatus that receives an image transmitted after a conversion different from a conversion performed on an image forming a moving image among signature conversions, and a receiving unit that receives image data; Discrete Fourier data creating means for creating discrete Fourier data from an image constituting the applied still image and an image constituting the transformed moving image which is a preceding image of the image, and the discrete Fourier data An image data receiving apparatus comprising: an image matching unit that compares and determines a spatial frequency of an image forming the still image with a previously registered image using 3. An image data transmitting and receiving system comprising: an image data transmitting device for transmitting images which are continuous in time; and an image data receiving device for receiving and processing the transmitted image data. The transmitting device includes: an image type determination unit configured to determine whether the target image is an image forming a moving image or a still image, based on a magnitude of a change amount between the target image and a previous image of the target image; First discrete means for performing a discrete cosine transform or a discrete sine transform on the target image determined to be an image constituting the image, and a discrete cosine transform for the target image determined to be the image constituting the still image. A second conversion unit that performs a conversion different from the conversion performed by the first conversion unit, out of the conversion and the discrete sine conversion, and transmits the target image converted by the first or second conversion unit. A receiving means for receiving image data transmitted from the transmitting apparatus; an image forming a converted still image; and an image forming the still image Means for creating discrete Fourier data from an image constituting a moving image which has been subjected to a transformation which is a previous image with respect to, and using the discrete Fourier data, an image constituting the still image is registered in advance. And an image matching means for comparing and judging the image. 4. An image acquiring means for acquiring an image which is temporally continuous, wherein one of the acquired images is set as a target image, and a target image is determined based on a magnitude of change between the target image and a previous image of the target image. Image type determining means for determining whether the image is a moving image or a still image, and a discrete cosine transform or a discrete sine transform on the target image determined to be a moving image. And a transform that is different from the transform performed by the first transform unit, of the discrete cosine transform and the discrete sine transform, on the target image determined to be an image forming a still image. A second conversion unit; and a transmission unit that transmits the target image converted by the first or second conversion unit. Recording medium. 5. A discrete cosine transform or a discrete sine transform is applied to an image constituting a moving image, which is a temporally continuous image, and a discrete cosine transform and a discrete cosine transform are applied to an image constituting a still image. A program for realizing, by a computer, a function of receiving an image transmitted after a conversion different from a conversion performed on an image forming a moving image among signature conversions, the reception being configured to receive image data Means for generating discrete Fourier data from an image constituting the transformed still image and an image constituting the transformed moving image which is a preceding image of the image constituting the still image. Fourier data creation means, and image matching means using the discrete Fourier data to compare and determine the target image with a pre-registered image, A computer-readable recording medium storing a program for causing revealed.