JP2005242592A - Image processing apparatus and image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus and an image processing method that correctly distinguish a subject included in a pickup image from the pickup image while reducing the number of dimensions of an image feature value and reducing computational complexity. <P>SOLUTION: An image processing server 1 comprises feature value extracting means for quantizing signals of a pickup image represented in a uniform color space in every arbitrary area of the pickup image, extracting frequencies of quantization level values on axes in the uniform color space as a color histogram, combining the color histogram of every arbitrary area, and extracting a feature value of the entire pickup image, and identifying means for distinguishing an unknown subject according to the feature value of the entire pickup image. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus and an image processing method for discriminating an unknown shooting target included in a shot image from a shot image.

2. Description of the Related Art Conventionally, there are various feature amounts used when determining a shooting target included in a captured image from a captured image. Examples include color feature values and shape (composition) feature values. In recent years, a method has been proposed in which a plurality of image feature amounts are combined to create a high-dimensional feature amount and a subject to be imaged is determined (see, for example, Patent Document 1).
Japanese Patent Laying-Open No. 2003-289551 (paragraphs “0089” to “0091”, FIG. 10)

 However, when the dimension of the feature quantity is increased by combining a plurality of image feature quantities, the amount of calculation processing increases according to the number of dimensions.

 In view of the above problems, the present invention is an image processing apparatus and an image processing method that can reduce the number of dimensions of an image feature amount, reduce the amount of calculation processing, and correctly determine an imaging target included in the captured image from the captured image. The purpose is to provide.

In order to achieve the above object, a first feature of the present invention is an image processing apparatus that discriminates an unknown shooting target included in a shot image from a shot image, and (a) for each arbitrary region of the shot image, First feature amount extraction means for quantizing a signal of a captured image expressed in a color space and extracting a frequency of a quantization level value of each axis in the color space as a color histogram; and (b) for each arbitrary region An image comprising: a second feature amount extraction unit that combines color histograms to extract a feature amount of the entire captured image; and (c) a determination unit that determines an unknown shooting target based on the feature amount of the entire captured image. The gist is that it is a processing device. Here, the “color space” includes uniform color space (L ^* , a ^* , b ^* space), RGB color space, CMYK color space, L ^* , u ^* , v ^* space, YUV space, XYZ space, etc. Various color spaces are included.

  According to the image processing apparatus according to the first feature, the number of dimensions of the image feature amount can be reduced, the amount of calculation processing can be reduced, and the shooting target included in the shot image can be correctly determined from the shot image. Here, the “number of dimensions” refers to the number of elements of the feature quantity that is a vector.

  The image processing apparatus according to the first feature further includes a weighting unit that multiplies the color histogram for each arbitrary region by an arbitrary weight, and the second feature amount extraction unit includes an arbitrary weighted multiplying unit. You may combine the color histogram for every area | region. According to this image processing apparatus, it is possible to balance the size of the feature vector extracted in each region.

  The arbitrary weight may be the reciprocal of the total value of the dimension values of the color histogram for each arbitrary region. According to this image processing apparatus, the vectors can be normalized and combined.

  In addition, the image processing apparatus according to the first feature includes a storage unit that stores a plurality of predetermined parameters calculated based on each feature amount of the entire plurality of captured images, and a parameter calculation unit that calculates the predetermined parameters. The determining unit may determine an unknown imaging target by using a predetermined parameter. According to this image processing apparatus, it is possible to discriminate an unknown imaging target using parameters calculated using the feature amount.

Also, l is the number of a known object, L is the number of known objects, N ^(l) is the number of images taken for the l-th known object, v is the number of feature dimensions, and V is the feature number. The maximum number of dimensions of the quantity, x ^(l) _{(n, v)} is the feature quantity of the vth dimension of the nth captured image belonging to the lth known imaging target, γ is an arbitrary value, θ _{(v, l )} Is a predetermined parameter in the v-th dimension of the feature quantity belonging to the l-th known imaging target, x ′ _(v) is a feature quantity in the v-th dimension of the captured image including the unknown imaging target, F _(l) As a discriminant value indicating the degree to which the unknown imaging target belongs to the l-th known imaging target, the parameter calculation means of the image processing apparatus according to the first feature calculates a predetermined parameter by the following equation:

The discriminating unit may recognize a photographed image having the largest discriminant value obtained by the following expression as a subject to be photographed.

  According to this image processing apparatus, it is possible to calculate a discrimination value using a parameter calculated using a feature amount and discriminate an unknown imaging target. Equations (1) and (2) are so-called NB (Naive Bayes) techniques, and the probability that an unknown subject is a known subject is calculated for each of a plurality of known subjects. And has an advantage of high processing speed.

  A second feature of the present invention is an image processing method for discriminating an unknown shooting target included in a shot image from a shot image, and (a) a shooting expressed in a color space for each arbitrary region of the shot image. A first feature amount extraction step for quantizing an image signal and extracting a frequency of a quantization level value of each axis in a color space as a color histogram, and (b) combining a color histogram for each arbitrary region and shooting It is an image processing method including a second feature amount extraction step for extracting a feature amount of the entire image, and (c) a step of determining an unknown shooting target based on the feature amount of the entire shot image. To do.

  According to the image processing method according to the second feature, it is possible to suppress the number of dimensions of the image feature amount, reduce the calculation processing amount, and correctly determine the shooting target included in the shot image from the shot image.

 According to the present invention, it is possible to provide an image processing apparatus and an image processing method that can suppress the number of dimensions of an image feature amount, reduce the amount of calculation processing, and correctly determine a shooting target included in the shot image from the shot image.

  Next, embodiments of the present invention will be described with reference to the drawings. In the following description of the drawings, the same or similar parts are denoted by the same or similar reference numerals. However, it should be noted that the drawings are schematic.

(Image processing system)
As illustrated in FIG. 1, the image processing system according to the present embodiment receives a user input, requests processing from the image processing server 1 (image processing apparatus), and outputs a processing result in the image processing server 1. Accepts processing requests from the terminal devices 2a, 2b, and 2c via the devices 2a, 2b, and 2c and the communication network (such as the Internet) 3, performs processing, and transmits processing results to the terminal devices 2a, 2b, and 2c. An image processing server 1.

  The user can use the terminal devices 2a, 2b, and 2c by switching between two modes, a learning mode and a discrimination mode. The “learning mode” is a mode in which the image processing server 1 learns a photographed image by photographing a known photographing target and transmitting the photographed image and information specifying the photographing target to the image processing server 1. . On the other hand, in the “discrimination mode”, an unknown shooting target is shot, and this shot image is transmitted to the image processing server 1, whereby the image processing server 1 determines the shooting target, and the determination result is displayed as the terminal device 2a. 2b and 2c.

  When the learning mode is set, the image processing server 1 performs grouping of the shooting targets from the information about the shooting targets and the shot images obtained by shooting the shooting targets received from the terminal devices 2a, 2b, and 2c. Remember.

  When it is in the discrimination mode, the image processing server 1 is registered in advance as a photographing target that is close to the position information of the terminal devices 2a, 2b, and 2c and the position where the captured images received from the terminal devices 2a, 2b, and 2c are photographed. Information is taken out, and a shooting target included in the shot image is determined probabilistically. In addition, the image processing server 1 transmits to the terminal devices 2a, 2b, and 2c a plurality of candidates for shooting targets that are the determination results, the probability that each candidate is a shooting target, and information related to these shooting targets.

  As shown in FIG. 2, the image processing server 1 includes a communication unit 11, a determination unit 12, a feature amount extraction unit 13, a registration unit 14, a learning unit 15, a calculation device 16, a storage device 17, Weighting means 18.

  The communication unit 11 receives the captured image and the information on the imaging target from the terminal devices 2a, 2b, and 2c via the communication network 3 (Internet or the like). Further, when in the discrimination mode, the communication unit 11 transmits information on the imaging target and the discrimination result to the terminal devices 2a, 2b, and 2c via the communication network 3 (Internet or the like).

  The feature amount extraction unit 13 extracts a feature amount that serves as an index when determining the shooting target. Specifically, the feature amount extraction unit 13 quantizes the signal of the captured image expressed in the uniform color space for each arbitrary region of the captured image, and the frequency of the quantization level value of each axis in the uniform color space. Are extracted as a color histogram. Then, the feature amount extraction means 13 combines the color histograms for each arbitrary region, and extracts the feature amount of the entire captured image. That is, the feature amount extraction unit 13 extracts the feature amount (x) of the entire captured image having elements for the number of dimensions (V) for each known captured image. The feature quantity (x) is a vector having a certain number of elements. In the following description, x and x 'are vectors.

Further, X represents a feature amount of a photographed image belonging to a plurality of known photographing objects. The feature quantity X is a matrix consisting of each element x ^(l) _{(n, v), where} ^l is the number of a known subject, n is the number assigned to each photographed image, and v is the dimension number of the feature quantity. expressed.

The learning unit 15 (parameter calculation unit) calculates the learning parameter Θ based on the feature quantity extracted by the feature quantity extraction unit 13 using the equation (1). The learning parameter Θ is a matrix in which V × L elements θ _{(v, l)} are arranged, where V is the maximum dimension of the feature quantity and L is the number of known shooting targets.

In Equation (1), l is the number of a known object, L is the number of known objects, N ^(l) is the number of images taken for the l-th known object, and v is the feature dimension number. , V is the maximum number of dimensions of the feature quantity, x ^(l) _{(n, v)} is the feature quantity of the vth dimension of the nth captured image belonging to the lth known imaging target, γ is an arbitrary value, θ _{(v, l)} is a learning parameter in the v-th dimension of the feature quantity belonging to the l-th known imaging target.

  In the above description, v, l, and n (lowercase) are variables, and V, L, and N (uppercase) are fixed values. X and x '(lower case) are vectors, and X and Θ (upper case) are matrices.

Specifically, the learning unit 15 (parameter calculation unit) calculates the v-th dimension learning parameter θ _{(v, l)} of the feature quantity belonging to the l-th known imaging target using the equation (1). To do.

The discriminating unit 12 discriminates an unknown imaging target by using the learning parameter calculated by the learning unit 15 (parameter calculating unit) using Expression (2) for the captured image received in the discrimination mode. .

In equation (2), x ′ _(v) is the feature quantity of the v th dimension of the captured image including the unknown imaging target, and F _(l) is the unknown imaging target belonging to the l th known imaging target. It is a discriminant value indicating the degree.

Specifically, the determination unit 12 belongs to a photographed image including an unknown photographing target using the learning parameter θ _{(v, l)} of the v th dimension of the feature amount belonging to the l th known photographing target. A value of F _(l) indicating the degree to which the feature quantity x ′ belongs to the l-th known imaging target is calculated. Here, it is determined that the larger the value of F _(l), the higher the probability that the unknown imaging target is the l-th known imaging target.

In this way, if F _(l) is calculated L times according to equation (2), according to the value of F _(l) , which of the L types of imaging targets the unknown imaging target belongs to is ranked. Can be determined.

  The registration unit 14 registers the feature amount and learning parameter for each captured image in the storage device 17 via the arithmetic device 16.

  The weighting means 18 multiplies the color histogram for each arbitrary region by an arbitrary weight. The feature amount extraction unit 13 combines the color histograms for each arbitrary region multiplied by the weight by the weighting unit 18. The arbitrary weight may be, for example, the reciprocal of the total value of the dimension values of the color histogram for each arbitrary region.

  The arithmetic device 16 controls the operations of the communication unit 11, the determination unit 12, the feature amount extraction unit 13, the registration unit 14, the learning unit 15, the storage device 17, and the weighting unit 18.

  The image processing server 1 according to the embodiment of the present invention includes a processing control device (CPU), and includes a communication unit 11, a determination unit 12, a feature amount extraction unit 13, a registration unit 14, a learning unit 15, and a weighting unit. 18 or the like as a module can be built in the CPU. These modules can be realized by executing a dedicated program for using a predetermined program language in a general-purpose computer such as a personal computer. In addition, the storage device 17 stores feature amounts of a plurality of captured images, a plurality of predetermined learning parameters calculated based on each of the plurality of captured images, captured image data, registration target information, registration target related information, determination values, and the like. A recording medium to be stored. Examples of the recording medium include RAM, ROM, hard disk, flexible disk, compact disk, IC chip, and cassette tape. According to such a recording medium, it is possible to easily store, transport, and sell captured image data, learning parameters, registration target information, and the like.

  The terminal devices 2a, 2b, and 2c can switch between two types of modes, a learning mode and a discrimination mode.

  In the learning mode, the terminal devices 2a, 2b, and 2c photograph a known subject to be photographed using the mounted camera. The terminal devices 2a, 2b, and 2c transmit the captured image to the image processing server 1 together with registration target information and registration target related information registered in advance by the user. Here, “registration target information” is information for specifying a shooting target in a shot image. Further, “registration target related information” includes position information of a shooting target, information (name, URL, etc.) related to the shooting target, and the like.

In the discrimination mode, the terminal devices 2a, 2b, and 2c photograph an unknown photographing target using the mounted camera. The terminal devices 2a, 2b, and 2c transmit the captured image to the image processing server 1 together with the position information. Then, the terminal devices 2a, 2b, and 2c receive a plurality of shooting target candidates that are the determination results from the image processing server 1, the probability that each candidate is a shooting target, and information related to the shooting targets, and Based on the value of F _(l) , “registration target information” and “registration target related information” are ranked and presented to the user. The user can easily extract desired data from the ranked candidates.

  As shown in FIG. 3, the terminal device 2 includes an input unit 21, a communication unit 22, an output unit 23, a photographing unit 24, a positioning unit 25, a calculation device 26, and a storage device 27.

  The communication unit 22 transmits the captured image and the information about the imaging target to the image processing server 1 via the communication network 3 (Internet). When the communication unit 22 is in the determination mode, the communication unit 22 receives information about the imaging target and the determination result from the image processing server 1 via the communication network 3 (Internet).

  Specifically, the imaging unit 24 refers to a mounted camera or the like, captures an object, and acquires a captured image.

  The positioning means 25 measures the position of the terminal device 2 and the position of the photographing target.

  The input means 21 refers to devices such as a touch panel, a keyboard, and a mouse. When an input operation is performed from the input means 21, corresponding key information is transmitted to the arithmetic device 26. The output means 23 refers to a screen such as a monitor, and a liquid crystal display (LCD), a light emitting diode (LED) panel, an electroluminescence (EL) panel, or the like can be used.

  The arithmetic device 26 controls operations of the input means 21, the communication means 22, the output means 23, the photographing means 24, the positioning means 25, and the storage device 27. The arithmetic unit 26 operates as a switching unit that switches between the learning mode and the discrimination mode based on key information input from the input unit 21.

  The storage device 27 is a recording medium that stores captured images, registration target information, registration target related information, and the like.

(Image processing method)
Next, the image processing method according to the present embodiment will be described with reference to FIGS.

  First, a method for registering captured images will be described with reference to FIG.

  (A) First, in step S101, the terminal device 2 captures a known registration target in the learning mode and acquires the image.

  (B) Next, in step S102, the terminal device 2 inputs registration target information, and in step S103, the terminal device 2 inputs registration target related information. The registration target information and registration target related information may be input in advance before shooting. For example, if the photographed image is a confectionery store as shown in FIG. 9, “cake shop” or the like is input as “registration target information”, and “AAA confectionery store” or AAA confectionery store as “registration target related information”. And the URL of the AAA pastry store.

  (C) Next, in step S <b> 104, the terminal device 2 acquires the position information, the positioning error, the shooting time, and the distance and direction to the shooting target if possible.

  (D) Next, in step S <b> 105, the terminal device 2 transmits registration target information, registration target related information, position information, and acquired image data to the image processing server 1.

  (E) Next, in step S106, the image processing server 1 transmits registration target information, registration target related information, position information, and an acquired image. In step S107, the image processing server 1 extracts the feature amount of the registered image. This feature amount extraction method will be described in detail later.

  (F) Next, in step S108, the image processing server 1 stores the registration target information, the registration target related information, the registration target image, the feature amount, and the registration time in the image processing server 1 in the storage device 17.

  Next, a method for learning an imaging target will be described with reference to FIG.

  (A) First, in step S201, the image processing server 1 reads shooting target information, shooting target related information, acquired image data, and feature amount from the storage device 17.

  (B) Next, in step S202, the image processing server 1 performs target grouping based on position information to narrow down the target. In the case where learning is performed as needed at the time of determination of an imaging target, which will be described later, position information of the terminal device 2 is received from the terminal device 2, and the imaging target within the search range is learned. Here, the “search range” refers to an area having a radius within (positioning error) + (distance to the object) with the discrimination target as the center. The position information used here can accept any position. For example, learning may be performed using a position where the determination is likely to be performed in advance, or learning may be performed as needed when performing determination using position information of the terminal.

  (C) Next, in step S203, the image processing server 1 performs target learning. Specifically, the learning parameter is calculated using the above-described equation (1).

  (D) Next, in step S204, the image processing server 1 stores the target learning result (learning parameter).

  Next, a method for determining an imaging target will be described with reference to FIG.

  (A) First, in step S301, the terminal device 2 captures an unknown discrimination target in the discrimination mode and acquires an image. Next, in step S302, the terminal device 2 acquires discrimination target related information (position information of the discrimination target, its positioning error, distance to the target, etc.).

  (B) Next, in step S303, the terminal device 2 transmits the discrimination target related information and the discrimination image to the image processing server 1. Next, in step S304, the image processing server 1 receives the discrimination target related information and the discrimination image. Next, in step S305, the image processing server 1 extracts the feature amount of the discrimination image. This feature amount extraction method will be described in detail later.

  (C) Next, in step S306, the image processing server 1 narrows down the discrimination target. Next, in step S307, the image processing server 1 determines whether learning of the search range is completed. If completed, the process proceeds to step S308. If not completed, the process proceeds to step S309.

  (D) In step S309, the image processing server 1 performs the learning described in step S203 in FIG. In step S308, the image processing server 1 probabilistically performs target discrimination using the feature amount of the discrimination image and the learning parameter. Specifically, a determination value is obtained using the above-described equation (2), and a captured image that is a discrimination candidate is obtained.

  (E) Next, in step S310, the image processing server 1 reads out the shooting target information and the shooting target related information related to the determination result candidate from the storage device 17, and obtains the shooting target information, the shooting target related information, and the determination result from the terminal device. 2 to send.

  (F) Next, in step S310, the terminal device 2 prioritizes the shooting target information and the shooting target related information based on the probabilistic discrimination result (for example, displays a high-probability candidate at the top of the screen). Present it to the user. As a result, the user can further extract information related to the target from the network based on an address such as a URL related to the unknown registration target.

  According to the determination processing shown in FIG. 6, for example, when the user images a store A that is an unknown image capturing object by the terminal device 2 and transmits the captured image to the image processing server 1, the image processing server 1 stores the store A. The name of the store A, the URL, etc. can be transmitted to the terminal device 2. For this reason, the terminal device 2 can easily obtain information regarding the subject to be photographed from the photographed image.

  Next, details of the feature amount extraction method in step S107 in FIG. 4 and step S305 in FIG. 6 will be described with reference to FIG.

  (A) First, in step S 401, the image processing server 1 acquires a target captured image using a camera mounted on the terminal device 2. Here, the case where the image shown in FIG. 9 is acquired will be described as an example. The acquired image may be subjected to general image correction such as white balance by the function of the camera or the terminal device 2.

  (B) Next, in step S402, the image processing server 1 performs noise correction image correction processing on the acquired image. In step S403, the captured image is divided into arbitrary regions with respect to the corrected image, and a color histogram for each region is extracted. The color histogram of each region is expressed as a Vc-dimensional feature amount (vector value having Vc elements). As a method of dividing the region, FIG. 9 shows a region divided into a lattice shape, but it may be a radial shape or a circular shape, and the divided shape is not limited thereto. Further, the area may not be equally divided. The method for extracting the color histogram will be described in detail later.

  (C) Next, in step S404, a color histogram for each arbitrary region is combined to extract a feature amount of the entire captured image. The feature amount is a vector value having a certain number of elements. For example, when the color histogram of each area is Vc dimension and the number of area divisions is S, combining is performed as an independent dimension so that the dimension V of the feature quantity obtained by combining both feature quantities is V = S × Vc. . For example, the photographed image shown in FIG. 9 is divided into 4 × vertical 6 × 24 = 24 regions. Therefore, if the color histogram of each region is 24 dimensions, the feature amount of the entire image region is 24 × 24 = 576. It becomes a dimension.

  Next, details of the color histogram extraction method in step S403 of FIG. 7 will be described with reference to FIG.

(A) First, in step S501, the image processing server 1 represents the corrected image in a uniform color space (L ^* , a ^* , b ^* ) that is a visually uniform space for each arbitrary region.

  (B) Next, in step S502, the image processing server 1 independently quantizes each axis in the uniform color space at equal intervals.

(C) Next, in step S503, the frequency of the quantization level values of L ^* , a ^* , and b ^* is set as a color histogram. For example, if the three-axis quantization level of L ^* , a ^* , and b ^* is 8, the color histogram has 24 (= 3 × 8) dimension (the Vc dimension described above). At this time, the color histogram includes the ratio of the three primary colors and the luminance value of each pixel. For example, the photographed image shown in FIG. 9 is divided into 4 × vertical 6 × horizontal = 24 areas, and therefore a 24-dimensional feature value is calculated for each of the 24 areas.

  (D) Next, in step S504, the color histogram for each arbitrary region is multiplied by an arbitrary weight. The arbitrary weight may be, for example, the reciprocal of the total value of the dimension values of the color histogram for each arbitrary region. Since each region is not necessarily equally divided, the color histogram can be normalized by multiplying the reciprocal of the total value as a weight. In addition, the weights to be multiplied such as multiplying a region close to the center of the photographed image are appropriately selected according to the actual situation.

(Action and effect)
According to the image processing apparatus and the image processing method according to the present embodiment, it is possible to extract a color histogram for each arbitrary region of a photographed image and determine a photographing target based on a feature amount obtained by combining them. For this reason, the number of dimensions of the image feature amount can be kept low, the amount of calculation processing can be reduced, and the shooting target included in the shot image can be correctly determined from the shot image.

For example, assume that the captured image shown in FIG. 9 is an image of 100 pixels × 100 pixels. A process for combining the luminance value and the color feature amount is performed on the captured image. Specifically, 100 × 100 = 10,000 dimensions are calculated using the luminance value of each pixel as a feature quantity, and the three-axis quantization levels of L ^* , a ^* , and b ^* are set to 8 as the color feature quantity. In this case, 3 × 8 = 24 dimensions are calculated. When these are combined, the characteristic amount of the entire photographed image becomes 10,000 + 24 = 10,024 dimensions. On the other hand, as described in the present embodiment, when the captured image shown in FIG. 9 is divided into 4 × 6 × 24 regions and a color histogram is extracted for each region, three axes L ^* , a ^* , and b ^* . When the quantization level of 8 is 8, the color histogram for each region has 24 (= 3 × 8) dimensions. When these are combined, the feature amount of the entire captured image is 24 dimensions × 24 areas = 576 dimensions. As described above, according to the present embodiment, it is possible to greatly reduce the number of dimensions of the image feature amount and reduce the calculation processing amount.

  Further, according to the image processing apparatus and the image processing method according to the present embodiment, a feature amount is extracted by multiplying a color histogram for each arbitrary region by an arbitrary weight and combining the weighted histograms. be able to. For this reason, it is possible to balance the size of the feature vector extracted in each region.

  Further, the arbitrary weight can be the reciprocal of the total value of the dimension values of the color histogram for each arbitrary region. For this reason, it is possible to perform the combination by normalizing the vectors.

  Further, according to the image processing apparatus and the image processing method according to the present embodiment, it is possible to discriminate an unknown shooting target by using a plurality of predetermined parameters calculated based on respective feature amounts of a plurality of shot images as a whole. Can do. For this reason, it is possible to determine an unknown imaging target using the parameter calculated using the feature amount.

  Further, according to the image processing apparatus and the image processing method according to the present embodiment, the predetermined parameter is calculated using the above-described equation (1), and the captured image having the largest discriminant value obtained by the equation (2) is obtained. Can be recognized as a shooting target. Expressions (1) and (2) are so-called NB (Naive Bayes) techniques, and the probability that an unknown imaging target corresponds to a known imaging object can be calculated for each of a plurality of known imaging objects. It is possible and has an advantage of high processing speed.

(Other embodiments)
Although the present invention has been described according to the above-described embodiments, it should not be understood that the descriptions and drawings constituting a part of this disclosure limit the present invention. From this disclosure, various alternative embodiments, examples and operational techniques will be apparent to those skilled in the art.

 For example, in the above-described embodiment, an example has been shown in which learning parameters are acquired using an NB (Naive Bayes) method to determine an unknown imaging target. However, not only NB, but also learning using other Bayesian methods and other learning algorithms such as SVM (Support Vector Machine), kNN (k Nearest Neighbor), LVQ (Learning Vector Quantization) You may acquire a parameter and discriminate | determine an unknown imaging | photography object.

 Further, the image processing server 1 according to the embodiment incorporates a communication unit 11, a determination unit 12, a feature amount extraction unit 13, a registration unit 14, a learning unit 15, and a weighting unit 18 in a CPU as a module. Although it has been described that the configuration can be made, they may be divided into two or more CPUs. In this case, it is assumed that the devices are connected by a bus or the like so that data can be exchanged between the CPUs.

  As described above, the present invention naturally includes various embodiments not described herein. Therefore, the technical scope of the present invention is defined only by the invention specifying matters according to the scope of claims reasonable from the above description.

1 is a configuration block diagram of an image processing system according to an embodiment of the present invention. It is a block diagram of the configuration of the image processing server according to the embodiment of the present invention. It is a block diagram of the configuration of a terminal device according to an embodiment of the present invention. It is a flowchart which shows the registration process which concerns on embodiment of this invention. It is a flowchart which shows the learning process which concerns on embodiment of this invention. It is a flowchart which shows the discrimination | determination process which concerns on embodiment of this invention. It is a flowchart which shows the feature-value extraction process which concerns on embodiment of this invention. It is a flowchart which shows the detail of step S403 of FIG. It is an example of the picked-up image which concerns on embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image processing server 2 Terminal device 11 Communication means 12 Object discrimination means 13 Feature quantity extraction means 14 Registration means 15 Learning means 16 Arithmetic device 17 Storage device 18 Weighting means 21 Input means 22 Communication means 23 Output means 24 Imaging means 25 Imaging means 25 Positioning means 26 Arithmetic device 27 Storage device

Claims (6)

An image processing apparatus for determining an unknown shooting target included in a captured image from a captured image,
First feature amount extraction for quantizing a signal of a photographed image expressed in a color space for each arbitrary region of the photographed image and extracting a frequency of a quantization level value of each axis in the color space as a color histogram Means,
A second feature amount extracting means for combining the color histograms for each arbitrary region and extracting the feature amount of the entire captured image;
An image processing apparatus comprising: a discriminating unit that discriminates the unknown imaging target based on a feature amount of the entire captured image. A weighting unit for multiplying the color histogram for each arbitrary region by an arbitrary weight;
The image processing apparatus according to claim 1, wherein the second feature amount extraction unit combines the color histograms of the arbitrary regions multiplied by weights.   The image processing apparatus according to claim 2, wherein the arbitrary weight is a reciprocal of a total value of dimension values of a color histogram for each arbitrary region. Storage means for storing a plurality of predetermined parameters calculated based on the respective feature amounts of the plurality of captured images;
Parameter calculating means for calculating the predetermined parameter;
The image processing apparatus according to claim 1, wherein the determination unit determines the unknown imaging target by using the predetermined parameter. l is the number of the known imaging target, L is the number of the known imaging targets, N ^(l) is the number of the captured images for the l-th known imaging target, v is the dimension number of the feature, and V is The maximum number of dimensions of the feature amount, x ^(l) _{(n, v)} is the feature amount of the vth dimension of the nth photographed image belonging to the lth known photographed object, γ is an arbitrary value, θ _{( v, l)} is the predetermined parameter in the v-th dimension of the feature quantity belonging to the l-th known imaging target, and x ′ _(v) is the feature quantity in the v-th dimension of the captured image including the unknown imaging target. , F _(l) as a discrimination value indicating the degree to which the unknown imaging target belongs to the l-th known imaging target,
The parameter calculation means calculates the predetermined parameter according to the following equation:

The discriminating unit discriminates the unknown imaging target based on the discriminant value obtained by the following expression.

The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus. An image processing method for determining an unknown shooting target included in a captured image from a captured image,
First feature amount extraction for quantizing a signal of a photographed image expressed in a color space for each arbitrary region of the photographed image and extracting a frequency of a quantization level value of each axis in the color space as a color histogram Steps,
A second feature amount extraction step of combining the color histograms for each arbitrary region and extracting the feature amount of the entire captured image;
An image processing method comprising: discriminating the unknown imaging target based on a feature amount of the entire captured image.

Images