JP2013218589A - Signal processor, signal processing method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a signal processor which solves the problem that two or more different featured values cannot be acquired form one video signal by one device.SOLUTION: The signal processor includes: an input signal acquisition unit for acquiring an input signal having two or more unit signals, the unit signal being a signal per predetermined unit quantity; an area division unit for dividing each of the two or more unit signals included in the input signal into two or more areas; a featured value acquisition unit for acquiring two or more different types of featured values one form each area of the two or more areas; and an output unit for outputting the featured values acquired by the featured value acquisition unit. Thus, each of the two or more unit signals included in the input signal is divided into two or more areas, and the different featured values are acquired from each area, thereby acquiring two or more different featured values from each unit signal.

Description

  The present invention relates to a signal processing apparatus for processing a signal.

  Conventionally, in order to detect a pull-down signal due to captions and edit points, a video signal processing apparatus has been developed that calculates an average of pixel value differences between frames or fields (see Patent Document 1).

JP2011-254392

  Normally, when calculating statistical features such as histograms from video signals in units of frames, etc., the number of features to be calculated and the types of features to be calculated will increase in proportion to the number of regions. Scale increases. However, a conventional video signal processing apparatus or the like cannot acquire two or more different feature amounts from one video signal without increasing the circuit scale.

  The signal processing apparatus according to the first aspect of the present invention includes an input signal acquisition unit that acquires an input signal having two or more unit signals that are signals per unit amount determined in advance, and two or more unit signals that the input signal has. A region dividing unit that divides a region into two or more regions, a feature amount acquiring unit that acquires two or more different types of feature amounts from each of the two or more regions, and a feature amount acquiring unit. A signal processing device including an output unit that outputs the feature amount.

  With such a configuration, two or more different feature quantities can be acquired from two or more unit signals of one input signal without increasing the circuit scale.

  The signal processing apparatus according to the second aspect of the invention integrates two or more feature quantities of the same type acquired by the feature quantity acquisition unit for each predetermined number of unit signals with respect to the first invention. The signal processing device further includes a feature amount integration unit, and the output unit outputs the feature amount integrated by the feature amount integration unit.

  With such a configuration, the accuracy of the acquired feature amount can be improved.

  In addition, the signal processing device according to the third aspect of the present invention acquires division pattern information indicating a pattern for dividing a unit signal into two or more areas and each of the two or more areas with respect to the first or second invention. A management information storage unit that stores two or more pieces of management information having two or more feature amount identifiers that identify feature amounts in association with each other; a management information acquisition unit that acquires one or more pieces of management information from the management information storage unit; The area dividing unit divides two or more unit signals into two or more areas according to the division pattern information included in the management information acquired by the management information acquiring unit, and the feature amount acquiring unit acquires the management information. This is a signal processing device that acquires, from each of the two or more areas, a feature quantity identified by the feature quantity identifier corresponding to each of the two or more areas.

  With this configuration, it is possible to switch between the division pattern and the feature amount acquired from each divided area for each unit signal.

  Further, the signal processing device according to the fourth aspect of the invention is a feature quantity identifier for storing two or more feature quantity identifiers for identifying feature quantities acquired from each of two or more areas, relative to the first or second invention. A storage unit; the region dividing unit divides each of the two or more unit signals into two or more predetermined regions; and the feature amount acquisition unit includes two or more predetermined units from the feature amount identifier storage unit. This is a signal processing device that acquires a feature amount identifier and acquires a feature amount identified by one of the feature amount identifiers from two or more regions in a predetermined order.

  With such a configuration, the feature amount acquired from each divided region can be switched for each unit signal.

  In addition, the signal processing apparatus according to the fifth aspect of the present invention obtains, from the third or fourth aspect, division pattern information indicating a pattern for dividing a unit signal into two or more areas and two or more areas. Feedback in which feedback management information having one or more pieces of management information associated with two or more feature amount identifiers that identify feature amounts and feature amount conditions that are conditions relating to one or more feature amounts in association with each other is stored. It is determined whether the feature information acquired by the management information storage unit and the feature amount acquisition unit or the feature amount integrated by the feature amount integration unit matches any one of the feature amount conditions included in the one or more feedback management information. One or more corresponding to the feature amount condition that the feature amount acquired by the feature amount acquisition unit from the feedback management information storage unit or the feature amount integrated by the feature amount integration unit matches when the determination unit and the determination unit match. The management information of And an area dividing unit that divides two or more unit signals into two or more areas according to the division pattern information included in the management information acquired by the management information acquiring unit, and obtains a feature amount acquiring unit. Is a feature amount identifier of the management information acquired by the management information acquisition unit, and is a signal processing device that acquires the feature amount identified by the feature amount identifier corresponding to each of the two or more regions from each of the two or more regions. is there.

  With such a configuration, it is possible to change the division pattern and the feature amount acquired from each divided region according to the acquired feature amount.

  The signal processing device according to the sixth aspect of the present invention is the feature amount for compressing the feature amount acquired by the feature amount acquisition unit or the feature amount integrated by the feature amount integration unit with respect to any one of the first to fifth inventions. The signal processing device further includes a compression unit, and the output unit outputs the feature amount compressed by the feature amount compression unit.

  With such a configuration, the data amount of the acquired feature amount can be reduced.

  According to the signal processing apparatus and the like according to the present invention, two or more different feature quantities can be acquired from two or more unit signals of one input signal without increasing the circuit scale.

Block diagram of signal processing apparatus 1 in the first embodiment A flowchart for explaining the overall operation of the signal processing apparatus 1 The figure which shows the example of the management information Image diagram showing an example of the same region segmentation and feature acquisition processing The figure which shows the example of the management information Image diagram showing an example of the same region segmentation and feature acquisition processing The figure which shows the example of the return management information Image diagram showing an example of the same region segmentation and feature acquisition processing The figure which shows the structural example of the signal processing circuit 2 The figure which shows the structural example of the signal processing circuit 3 The figure which shows the structural example of the signal processing circuit 3 Overview of the computer system in the above embodiment Block diagram of a computer system in the above embodiment

  Hereinafter, embodiments of a signal processing apparatus and the like according to the present invention will be described with reference to the drawings. In addition, since the component which attached | subjected the same code | symbol in embodiment performs the same operation | movement, description may be abbreviate | omitted again. In addition, in a storage unit or the like in which predetermined information can be stored, the process in which information is stored in the storage unit does not matter. For example, information may be stored in a storage unit or the like via a recording medium. Information transmitted via a communication line or the like may be stored in a storage unit or the like. Furthermore, information input via the input device may be stored in a storage unit or the like. In addition, the format, content, and the like of each information described in this embodiment are merely examples, and the format, content, and the like are not limited as long as the meaning of each information can be indicated.

(Embodiment 1)
In the present embodiment, a signal processing device that divides an input signal into two or more regions, acquires two or more different feature amounts from the two or more regions, and outputs them to an external device or the like will be described.

  FIG. 1 is a block diagram of a signal processing apparatus 1 in the present embodiment. The signal processing apparatus 1 includes a management information storage unit 101, a feature amount identifier storage unit 102, a feedback management information storage unit 103, an input signal acquisition unit 104, a determination unit 105, a management information acquisition unit 106, a region division unit 107, and a feature amount acquisition. Unit 108, feature amount integration unit 109, feature amount compression unit 110, and output unit 111.

  The management information storage unit 101 associates division pattern information indicating a pattern for dividing each of two or more unit signals included in an input signal into two or more regions and two or more feature quantity identifiers for identifying feature quantities. Two or more pieces of management information are stored.

  The input signal is a signal acquired by an input signal acquisition unit 104 described later. The input signal is usually a video signal obtained by converting a video signal, but may be an audio signal obtained by converting a sound signal, a composite signal obtained by converting a video signal and a sound signal, or the like. Also, the input signal usually has two or more unit signals. The unit signal is a signal per unit amount determined in advance. The unit amount is usually a frame, but may be a second. That is, the unit signal is usually an input signal per frame, but may be an input signal per second. When the input signal is a video signal, the unit signal is an image.

  A region is a part of a unit signal. In addition, the shape of the region is usually rectangular, but may be other shapes. In addition, the area is hereinafter referred to as a divided area as appropriate.

  The division pattern information is information having one or more area information that is information indicating the division area. The area information is usually coordinates on the image. The coordinates include, for example, coordinates indicating the starting point of the divided area (rectangular upper left coordinates), coordinates indicating the ending point of the divided area (rectangular lower right coordinates), coordinates indicating the intersection of the divided areas, and dividing. Coordinates indicating the center of the area. Further, the area information may be time. The time includes, for example, a time indicating the start position of the divided area, a time indicating the end position of the divided area, a time indicating the boundary between the divided areas, and a time indicating the center of the divided area. Further, the division pattern information may be a calculation formula for calculating coordinates indicating the division area. The calculation formula includes a variable indicating that, for example, an image size (dimension), an audio bit rate, or other numerical values are substituted. Further, the division pattern information may indicate that the unit signal is not divided into two or more areas. The division pattern information in this case is, for example, “division: none”.

  The feature amount is, for example, a pixel histogram, adjacent pixel difference histogram, inter-frame difference histogram, pixel value, luminance value, frequency distribution, dynamic range, spatiotemporal correlation, noise amount, pixel average value, pixel variance value, etc. is there.

  The feature amount identifier is information for identifying feature amounts acquired from each of two or more divided regions. The feature quantity identifier includes, for example, the name of the feature quantity, the name of the program that obtains the feature quantity, the program that obtains the feature quantity, the name of the circuit that obtains the feature quantity, the configuration of the circuit that obtains the feature quantity, and the feature quantity. This is a program for rewriting the configuration of a circuit to be acquired. The feature amount identifier may indicate that the feature amount is not acquired. The feature quantity identifier in this case is, for example, “feature quantity: none”.

  The size of the area indicated by the area information associated with the same two or more feature quantity identifiers may be the same or different.

  The feature quantity identifier storage unit 102 stores two or more feature quantity identifiers for identifying feature quantities.

  The feedback management information storage unit 103 stores feedback management information having a feature quantity condition that is a condition relating to one or more feature quantities and one or more management information in association with each other. The feedback management information is information for changing the pattern for dividing the unit signal and the feature amount acquired from each of the two or more divided regions according to the acquired feature amount. The feedback management information is information used by the determination unit 105 described later.

  Further, the feature amount condition includes, for example, (1) a condition related to a pixel histogram, (2) a condition related to an adjacent pixel difference histogram, and (3) a condition related to an inter-frame difference histogram. (1) is, for example, “the ratio of pixels lower than a certain value is equal to or higher than a certain ratio”, “the ratio of pixels higher than a certain value is equal to or higher than a certain ratio”, and the like. Further, (2) is, for example, “the ratio of pixels having a difference equal to or greater than a certain value is equal to or greater than a certain ratio” or “the ratio of pixels having a difference equal to or smaller than a certain value is equal to or less than a certain ratio And so on. Further, (3) is the same as (2), for example. The feature amount condition may be a combination of the above conditions (1) to (3). The combination is a logical product (AND) or a logical sum (OR).

  The input signal acquisition unit 104 acquires an input signal. The acquisition includes reception by broadcasting, reception by communication, reading from a recording medium, and the like. Further, the input signal acquisition unit 104 normally accumulates unit signals included in the input signal sequentially in a predetermined storage area. The input signal acquisition unit 104 is usually realized by a means for receiving a broadcast, a means for reading data, and the like.

  The determination unit 105 includes a feature amount integrated by a feature amount integration unit 109 (to be described later) (hereinafter referred to as an integrated feature amount), which is one of the feature amount conditions of one or more feedback management information stored in the feedback management information storage unit 103. It is determined whether or not they match. Further, the determination unit 105 may determine whether a feature amount acquired by a feature amount acquisition unit 108 (to be described later) (hereinafter, acquired feature amount) meets any of the feature amount conditions. Further, the determination unit 105 may determine whether or not a feature amount (hereinafter, compressed feature amount) compressed by a feature amount compression unit 110 (to be described later) meets any one of the feature amount conditions.

  For example, the feature amount condition is “(pixel value <20)> 70%” indicating that the ratio of pixels having a pixel value lower than 20 in the pixel histogram is 70% or more. It is assumed that either the feature amount or the compressed feature amount is “(pixel value <20) = 80%” indicating that the ratio of pixels having a pixel value lower than 20 in the pixel histogram is 80%. In this case, the determination unit 105 determines whether or not the feature amount matches the feature amount condition, and determines that they match.

  The management information acquisition unit 106 acquires one or more management information from the management information storage unit 101. Here, the area dividing unit 107 described later normally divides one unit signal into two or more areas sequentially. Therefore, the management information acquisition unit 106 normally acquires one management information before the division. The “before division” is usually immediately before division. The area dividing unit 107 described later normally divides a unit signal for each predetermined number of unit signals. That is, the area dividing unit 107 described later normally divides the unit signal for each predetermined period. The period is usually the number of frames, but may be the number of seconds. In this case, the area dividing unit 107 divides two or more unit signals in the one cycle. Therefore, the management information acquisition unit 106 may acquire, for example, the same number of pieces of management information as the number of unit signals that the area division unit 107 described later divides in one cycle for each cycle. As described above, the management information acquisition unit 106 only needs to be able to acquire the management information before the area division unit 107 described later divides the unit signal into two or more areas, and the acquisition timing is not limited.

  Further, the management information acquisition unit 106 may acquire the management information in the order stored in the management information storage unit 101, for example, or may acquire at random regardless of the stored order. That is, the management information acquisition unit 106 only needs to acquire one or more pieces of management information from the management information storage unit 101, and the acquisition order is not limited.

  For example, it is assumed that an area dividing unit 107 described later divides a first unit signal included in an input signal. In this case, the management information acquisition unit 106 acquires the first management information stored in the management information storage unit 101, for example. Further, for example, it is assumed that a region dividing unit 107 described later divides a second unit signal included in the input signal. In this case, the management information acquisition unit 106 acquires the second management information stored in the management information storage unit 101, for example.

  Further, for example, it is assumed that an area dividing unit 107 described later divides three unit signals in one cycle. In this case, for example, the management information acquisition unit 106 acquires three management information from the management information storage unit 101 for each period. That is, in this case, the management information acquisition unit 106 acquires management information three by three.

  That is, the management information acquisition unit 106 sequentially manages one piece of management information corresponding to each unit signal divided into two or more regions by the region dividing unit 107 described later in accordance with the division timing. 101.

  In addition, when the determination unit 105 determines that a feature amount such as an integrated feature amount matches any one of one or more feature amount conditions, the management information acquisition unit 106 normally receives a unit signal from the region division unit 107 described later. Before the division, one or more pieces of management information corresponding to a feature amount condition that the integrated feature amount and the like match are acquired from the feedback management information storage unit 103. Since “before division” has already been described, description thereof is omitted. The feature amount condition is a feature amount condition included in the feedback management information stored in the feedback management information storage unit 103.

  That is, when the determination unit 105 determines that the feature amount meets one of the feature amount conditions of 1 or more, the management information acquisition unit 106 determines the unit signal that is divided into two or more regions by the region division unit 107 described later. One management information corresponding to each is sequentially acquired from the feedback management information storage unit 103 in accordance with the division timing.

  The area dividing unit 107 divides two or more unit signals included in the input signal into two or more areas. The “dividing” refers to acquiring area information indicating a divided area created by the division, associating division pattern information used in the division with each unit signal, and the like.

  The input signal acquisition unit 104 normally accumulates unit signals included in the acquired input signal sequentially in a predetermined storage area. Therefore, the area dividing unit 107 sequentially acquires the accumulated unit signals and divides the unit signals into two or more areas.

The area dividing unit 107 performs the following division, for example.
(1) Fixed division (2) Variable division (3) Division using feedback

  (1) is to divide any unit signal with a predetermined division pattern. In this case, the division pattern information indicating the division pattern is usually stored in an arbitrary storage area or a predetermined area.

  Further, (2) is management information acquired by the management information acquisition unit 106, which divides the unit signal according to the division pattern information included in the management information corresponding to each unit signal. That is, (2) is to select and divide a division pattern for each unit signal. In this case, the management information acquired by the management information acquisition unit 106 is management information stored in the management information storage unit 101.

  (3) is to divide the unit signal in accordance with a feature quantity such as an integrated feature quantity. Specifically, (3) is the management information acquired by the management information acquisition unit 106 as in (2), and divides the unit signal according to the division pattern information included in the management information corresponding to each unit signal. That is. In this case, the management information acquired by the management information acquisition unit 106 is management information stored in the feedback management information storage unit 103.

  In addition, for example, when the division pattern information used for dividing the unit signal is a calculation formula, the area division unit 107 substitutes a numerical value corresponding to the variable for a variable included in the calculation formula, and creates a new division pattern. Generate information. Then, the area dividing unit 107 divides the unit signal using the generated division pattern information.

  For example, the division pattern information is “{(1, 1) (1920, 360)}, {(1, 361) (1920, 720)}, {(1, 721) (1920, 1080)}”. . The division pattern information means that each set of coordinates is divided into regions having a rectangular upper left coordinate and a lower right coordinate. Therefore, in this case, the area dividing unit 107 sets the unit signals to “(1, 1) (1920, 360)”, “(1, 361) (1920, 720)”, “(1, 721) (1920, 1080)”. ) ”. For example, assume that the division pattern information is “(960, 540)”. The division pattern information means that the coordinates are divided into regions where the boundary lines are the intersections between the divided regions. Therefore, in this case, the area dividing unit 107 converts the unit signals into, for example, “(1, 1) (960, 540)”, “(961, 1) (1920, 540)”, “(1, 541) ( 960, 1080) ”and“ (961, 541) (1920, 1080) ”.

  Further, for example, the division pattern information is “{(1,1) (W, H / 3)}, {(1, H / 3 + 1) (W, 2H / 3)}, {(1,2H / 3 + 1) ( W, H)} ”. In this division pattern information, “W” is a variable indicating that the width (horizontal length) of the image is substituted, and “H” is that the height (vertical length) of the image is substituted. It is a variable to indicate. Therefore, in this case, the region dividing unit 107 substitutes the width and height of the image indicated by the unit signal to be divided into variables included in the division pattern information, and generates new division pattern information. Here, assuming that the width of the image to be divided is “1920” and the height is “1080”, the area dividing unit 107 displays the division pattern information “{(1, 1) (1920, 360)”. }, {(1,361) (1920,720)}, {(1,721) (1920,1080)} ".

  The feature quantity acquisition unit 108 acquires two or more feature quantities from each of the two or more divided areas. The area dividing unit 107 normally divides two or more unit signals sequentially into two or more areas. Therefore, the feature quantity acquisition unit 108 sequentially acquires the divided unit signals, and acquires two or more feature quantities from each of the two or more divided areas of the unit signal. Since the method and procedure for acquiring the feature amount from the input signal are known, detailed description is omitted.

  Further, the feature amount acquisition unit 108 may acquire the feature amounts so that the feature amounts acquired from each of the two or more divided regions are all different. In addition, the feature amount acquisition unit 108 may acquire feature amounts so that, for example, the feature amounts acquired from adjacent divided regions are different. In this case, the feature quantity acquisition unit 108 may acquire the same type of feature quantity from two or more divided regions.

In addition, the feature amount acquisition unit 108 performs the following acquisition, for example.
(1) Fixed acquisition (2) Variable acquisition (3) Acquisition using feedback

  (1) is to acquire two or more feature quantities of the same type from any unit signal. In this case, the area dividing unit 107 divides each unit signal according to the same division pattern information. For example, it is assumed that the area dividing unit 107 divides all unit signals into three areas identified by α, β, and γ. In this case, the feature amount acquisition unit 108 acquires, for example, three feature amounts identified by A, B, and C from each of the three regions of all unit signals.

  In this case, the feature quantity acquisition unit 108 is a feature quantity identifier corresponding to the division pattern information used by the area dividing unit 107, and the feature quantity identified by the feature quantity identifier corresponding to each of the two or more divided areas. , From each of the two or more divided areas. Further, the division pattern information and the feature amount identifier are normally stored in an arbitrary storage area or a predetermined area, but may be stored in the management information storage unit 101.

(2) is to acquire two or more feature amounts while switching the feature amounts to be acquired from each divided region. (2) can be further divided into the following two acquisitions according to the method of dividing the region.
(A) Acquisition that switches only the feature amount to be acquired (b) Acquisition that switches the feature amount to be acquired according to the division of the region

  (A) is to acquire the same type of feature quantity from each unit signal, but to switch the feature quantity to be obtained from each divided region for each unit signal. In this case, the area dividing unit 107 divides each unit signal according to the same division pattern information. For example, it is assumed that the region dividing unit 107 divides each unit signal into three regions identified by α, β, and γ. The feature quantity acquisition unit 108 acquires three feature quantities identified by A, B, and C from each unit signal. In such a case, for example, the feature amount acquisition unit 108 first acquires A from I, B from β, and C from γ for the first unit signal. Then, the feature amount acquisition unit 108 acquires B from α, C from β, and A from γ for the second unit signal. Then, the feature amount acquisition unit 108 acquires C from α, A from β, and B from γ for the third unit signal. Then, the feature quantity acquisition unit 108 acquires a feature quantity for the fourth unit signal, similarly to the first unit signal.

  As described above, in the case of (a), for example, the feature amount is acquired from each of the two or more divided regions while switching the combination of the divided region and the feature amount for each period. In this case, for example, the feature amount acquisition unit 108 acquires one or more feature amount identifiers from the feature amount identifier storage unit 102 and associates them with two or more regions divided by the region dividing unit 107 according to a predetermined rule. . At this time, the feature quantity acquisition unit 108 usually acquires the same number of feature quantity identifiers as the divided areas. Then, the feature amount acquisition unit 108 acquires the feature amount identified by the corresponding feature amount identifier from each of the divided areas.

  Further, (b) is to acquire two or more feature amounts of the same type from any unit signal, but to switch the division pattern of the region and the feature amount acquired from each divided region for each unit signal. . In this case, the area dividing unit 107 divides each unit signal into two or more areas according to the division pattern information corresponding to each unit signal. For example, the management information acquisition unit 106 acquires division pattern information indicating division into three regions identified by α, β, and γ, and three feature amounts A, B, and C from the respective regions. It is assumed that management information having a feature quantity identifier indicating that is acquired. Then, it is assumed that the area dividing unit 107 divides the unit signal corresponding to the management information into the three areas according to the division pattern information included in the management information. Then, the feature amount acquisition unit 108 acquires A from I, B from β, and C from γ for the unit signal.

  In addition, for example, the management information acquisition unit 106 divides into four areas identified by α, β, γ, and δ, and three pieces of A, B, and C from the divided areas. It is assumed that management information having a feature amount identifier indicating acquisition of a feature amount is acquired. Then, it is assumed that the area dividing unit 107 divides the unit signal corresponding to the management information into the four areas according to the division pattern information included in the management information. Then, the feature amount acquisition unit 108 acquires A from I, B from β, C from γ, and A from δ for the unit signal.

  As described above, in the case of (b), for example, the feature quantity is acquired from each of the two or more divided areas while switching the combination of the division pattern information and the feature quantity acquired from each divided area for each period. To do. In this case, the feature amount acquisition unit 108 is typically a feature amount identifier included in the management information acquired by the management information acquisition unit 106, and the feature amount identified by the feature amount identifier corresponding to each of the two or more divided regions is Obtained from each of the two or more divided regions.

  Further, (3) is to switch the feature amount acquired from each divided region according to the feature amount such as the integrated feature amount. Specifically, (3) is a feature amount identifier included in the management information acquired by the management information acquisition unit 106 as in (2), and is identified by a feature amount identifier corresponding to each of two or more divided regions. Is obtained from each of the two or more divided regions. In this case, the management information acquired by the management information acquisition unit 106 is management information stored in the feedback management information storage unit 103.

  The feature amount acquisition unit 108 can be usually realized by an MPU, a memory, or the like. The processing procedure of the feature amount acquisition unit 108 is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit). When realized by hardware, the feature amount acquisition unit 108 can be realized by an LSI such as FPGA (Field-Programmable Gate Array) capable of dynamic reconfiguration, for example.

  The feature amount integration unit 109 integrates two or more acquired feature amounts of the same type for each predetermined unit signal period. “Same type” means the same type and may indicate the same characteristics. Further, “integrating” means calculating an average, obtaining the largest number, adding up, and the like. For example, when the period is “3 frames”, the feature amount integration unit 109 integrates the same type of feature amounts acquired by the feature amount acquisition unit 108 every three frames. The feature amount integration unit 109 may integrate the acquired feature amounts compressed by the feature amount compression unit 110 described later.

  The feature amount compression unit 110 compresses the acquired feature amount or the integrated feature amount. “Compressing” means reducing the amount of data. The compression is usually reversible compression. The compression method is not limited.

  The output unit 111 outputs an acquired feature value, an integrated feature value, or a compressed feature value. The output is a concept including transmission to an external device, accumulation in a recording medium, delivery of a processing result to another processing device or another program, and the like.

  The management information storage unit 101, the feature amount identifier storage unit 102, and the feedback management information storage unit 103 are preferably non-volatile recording media, but can also be realized by volatile recording media.

  In addition, the determination unit 105, the management information acquisition unit 106, the region division unit 107, the feature amount integration unit 109, the feature amount compression unit 110, and the output unit 111 can be usually realized by an MPU, a memory, or the like. The processing procedure of the determination unit 105 or the like is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit). When realized by hardware, the determination unit 105 and the like can be realized by an integrated circuit such as an LSI, for example.

  Next, the overall operation of the signal processing apparatus 1 will be described using the flowchart of FIG. In the flowchart of FIG. 2, it is assumed that n feedback management information is stored in the feedback management information storage unit 103.

  (Step S201) The area dividing unit 107 determines whether a unit signal is stored in a predetermined storage area. If it is stored, the process proceeds to step S202; otherwise, the process returns to step S201. Note that the unit signal is a unit signal that the input signal acquired by the input signal acquisition unit 104 has, and is a unit signal that the input signal acquisition unit 104 sequentially accumulates in a predetermined storage area.

  (Step S202) The area dividing unit 107 acquires a unit signal from a predetermined storage area.

  (Step S203) The determination unit 105 determines whether the feature amount integration unit 109 has integrated the feature amounts. If they are integrated, the process proceeds to step S215, and if not, the process proceeds to step S204.

  (Step S204) The management information acquisition unit 106 acquires predetermined management information from the management information storage unit 101.

  (Step S205) The area dividing unit 107 divides the unit signal acquired in step S202 into two or more areas according to the division pattern information included in the management information acquired in step S204. Here, it is assumed that the unit signal is divided into m areas.

  (Step S206) The feature quantity acquisition unit 108 sets 1 to the counter i.

  (Step S207) The feature quantity acquisition unit 108 is a feature quantity identifier included in the management information acquired in step S204 from the i-th area (hereinafter, area [i]) of the unit signal, and corresponds to the area [i]. The feature quantity identified by the feature quantity identifier is acquired from the area [i].

  (Step S208) The feature quantity acquisition unit 108 accumulates the feature quantity of the area [i] acquired in step S207 in a predetermined storage area.

  (Step S209) The feature quantity acquisition unit 108 determines whether i is m. When it is m, it progresses to step S211, and when that is not right, it progresses to step S210.

  (Step S210) The feature quantity acquisition unit 108 increments i by 1, and returns to step S206.

  (Step S211) The feature quantity integration unit 109 determines whether it is the timing of feature quantity integration. For example, the feature amount integration unit 109 counts the number of unit signals divided by the region division unit 107 and determines whether the number matches a predetermined cycle. If it is timing, the process proceeds to step S212, and if not, the process returns to step S201.

  (Step S212) The feature amount integration unit 109 acquires feature amounts from a predetermined storage area and integrates them for each feature amount of the same type.

  (Step S213) The feature value compression unit 110 compresses the feature values integrated in step S212.

  (Step S214) The output unit 111 outputs the feature amount compressed in step S213.

  (Step S215) The management information acquisition unit 106 sets 1 to the counter j.

  (Step S216) The management information acquisition unit 106 determines whether or not the integrated feature amount matches the feature amount condition [j] included in the feedback management information [j] stored in the feedback management information storage unit 103. If they match, the process proceeds to step S217; otherwise, the process proceeds to step S218.

  (Step S217) The management information acquisition unit 106 acquires the management information [j] included in the feedback management information [j] from the feedback management information storage unit 103.

  (Step S218) The management information acquisition unit 106 determines whether j is n. When it is n, it progresses to step S204, and when that is not right, it progresses to step S219.

  (Step S219) The management information acquisition unit 106 increments j by 1.

  In the flowchart of FIG. 2, the process may be terminated by powering off or a process termination interrupt.

(Concrete example)
Next, a specific example of the operation of the signal processing device 1 will be described. In this specific example, it is assumed that the signal processing device 1 is a circuit that performs signal processing (hereinafter, a signal processing circuit). The input signal acquired by the input signal acquisition unit 104 is a video signal.

(Example 1)
In this example, it is assumed that the management information storage unit 101 stores the management information shown in FIG. The management information includes an ID (item name: ID) for uniquely identifying a record, two or more division pattern information (item name: region), two or more feature quantity identifiers (item name: feature quantity), and Have The management information acquisition unit 106 acquires one piece of management information from the management information storage unit 101 every time the area division unit 107 divides a unit signal. Further, the acquisition procedure is in the order of IDs, and it is assumed that the acquisition is repeated in that order.

  First, the input signal acquisition unit 104 receives a video signal. Then, the input signal acquisition unit 104 sequentially accumulates unit signals (images) included in the video signal in a predetermined storage area. Note that the storage area is usually a so-called queue, and a unit that acquires stored unit signals acquires unit signals in the order in which they are stored.

  Next, the area dividing unit 107 acquires the first image from a predetermined storage area. Then, the management information acquisition unit 106 acquires management information “ID = 001” from the management information of FIG. 3. The management information is obtained by dividing the image into “(1, 1) to (360, 360), (361, 1) to (1920, 360), (1, 361) to (1920, 720), ( 1, 721) to (1920, 1080) ", and a pixel histogram, non-acquisition, inter-frame pixel difference histogram, and adjacent pixel difference histogram from each of the regions by the feature amount identifier. The four feature quantities are acquired. “Non-acquisition” means that no feature value is acquired.

  Next, the area dividing unit 107 divides the acquired image into the above four areas according to the division pattern information included in the management information “ID = 001” in FIG. 3 acquired by the management information acquiring unit 106. Here, it is assumed that the four regions are region A, region B, region C, and region D in the order of the above-described division pattern information.

  Next, the feature amount acquisition unit 108 acquires a pixel histogram from the region A, an inter-frame pixel difference histogram from the region C, an adjacent pixel difference histogram from the region D, and acquires nothing from the region B. Then, the feature quantity acquisition unit 108 accumulates the acquired feature quantity in a predetermined storage area.

  Thereafter, the area dividing unit 107 sequentially acquires the next image from a predetermined storage area. Then, for each acquisition, the management information acquisition unit 106 sequentially acquires “ID = 002”, “ID = 003”, “ID = 001”, “ID = 002”,... Get management information. Then, the area dividing unit 107 divides the image into two or more areas in accordance with the division pattern information included in the acquired management information, and the feature amount acquiring unit 108 uses the management information acquired from each area. The feature quantity identified by the feature quantity identifier shown to correspond to is acquired.

  The state of the above processing is, for example, FIG. In this example, the region dividing unit 107 and the feature amount acquiring unit 108 divide the image into two or more regions using the management information acquired by the management information acquiring unit 106 as shown in FIG. A feature amount identifier is associated with each region, and a feature amount identified by the feature amount identifier corresponding to each region is acquired from the two or more regions.

  Next, the feature amount integration unit 109 integrates the four feature amounts acquired as described above for each type. Then, the output unit 111 outputs the integrated four feature amounts.

(Example 2)
In this example, it is assumed that the management information storage unit 101 stores the management information shown in FIG. The management information includes an ID (item name: ID) for uniquely identifying a record, two or more division pattern information (item name: region), two or more feature quantity identifiers (item name: feature quantity), and Have In addition, the management information acquisition unit 106 acquires management information from the management information storage unit 101 for each period in which the region division unit 107 and the feature amount acquisition unit 108 divide unit signals and acquire feature amounts. Shall. The period is the number of management information stored in the management information storage unit 101. That is, the management information acquisition unit 106 acquires all the management information in FIG. 5 every time the area division unit 107 processes three images. Also, the description of the same processing as the processing described in Example 1 is omitted.

  First, the area dividing unit 107 acquires the first image from a predetermined storage area. Then, the management information acquisition unit 106 acquires all the management information in FIG.

  Next, the area dividing unit 107 divides the acquired image into two areas according to the division pattern information included in the management information “ID = 001” in FIG. 5 acquired by the management information acquiring unit 106. Here, it is assumed that the two areas are an area A and an area B in the order of the division pattern information of “ID = 001” in FIG.

  Next, the feature amount acquisition unit 108 acquires a pixel histogram from the region A and an adjacent pixel difference histogram from the region B. Then, the feature quantity acquisition unit 108 accumulates the acquired feature quantity in a predetermined storage area.

  Thereafter, the area dividing unit 107 sequentially acquires the next image from a predetermined storage area. Then, the area dividing unit 107 divides the image into two or more areas in accordance with the division pattern information included in the management information acquired by the management information acquiring unit 106, and the feature amount acquiring unit 108 acquires from each of the areas. The feature amount identified by the feature amount identifier indicated by the management information corresponding to each area is acquired.

  The state of the above processing is, for example, FIG. In this example, the region dividing unit 107 and the feature amount acquiring unit 108 divide the image into two or more regions using the management information acquired by the management information acquiring unit 106 as shown in FIG. A feature quantity identifier is associated with each area, and a feature quantity identified by the feature quantity identifier corresponding to each area is acquired from each of the two or more areas.

(Example 3)
In this example, it is assumed that the management information storage unit 101 stores the management information shown in FIG. Further, it is assumed that the feedback management information storage unit 103 stores the feedback management information shown in FIG. The feedback management information includes an ID (item name: ID) for uniquely identifying a record, one feature amount condition (item name: condition), one or more division pattern information (item name: region), It has one or more feature quantity identifiers (item name: feature quantity). Note that the division pattern information and the feature amount identifier are management information.

  First, for example, as in Example 1, it is assumed that the feature amount integration unit 109 integrates four feature amounts, and the feature amount compression unit 110 compresses each of the four feature amounts.

  Next, it is assumed that the determination unit 105 determines that, among the four feature amounts, the pixel histogram matches the feature amount condition of “ID = 012” in FIG. The feature amount condition indicates that the number of pixels having a pixel value of 30 or less is 70% or more of the number of pixels of the processing target image.

  Next, the management information acquisition unit 106 acquires the management information of “ID = 012” in FIG. The area dividing unit 107 does not divide the acquired image in accordance with the division pattern information “ID = 012” in FIG. 7 acquired by the management information acquiring unit 106. Then, the feature amount acquisition unit 108 acquires a pixel histogram from the image acquired by the region division unit 107.

  The state of the above processing is, for example, FIG. In this example, the area dividing unit 107 and the feature amount acquisition unit 108, as shown in FIG. 8, use the management information acquired from the management information storage unit 101 acquired by the management information acquisition unit 106 to divide the image into two or more areas. And the feature amount is acquired from each of the two or more regions. Thereafter, the area dividing unit 107 and the feature amount acquisition unit 108 acquire the feature amount from the image using the management information acquired by the management information acquisition unit 106 from the feedback management information storage unit 103.

  First, the area dividing unit 107 acquires the first image from a predetermined storage area. Then, the management information acquisition unit 106 acquires all the management information in FIG.

  Next, the area dividing unit 107 divides the acquired image into two areas according to the division pattern information included in the management information “ID = 001” in FIG. 5 acquired by the management information acquiring unit 106. Here, it is assumed that the two areas are an area A and an area B in the order of the division pattern information of “ID = 001” in FIG.

  Next, the feature amount acquisition unit 108 acquires a pixel histogram from the region A and an adjacent pixel difference histogram from the region B. Then, the feature quantity acquisition unit 108 accumulates the acquired feature quantity in a predetermined storage area.

  Thereafter, the area dividing unit 107 sequentially acquires the next image from a predetermined storage area. Then, the area dividing unit 107 divides the image into two or more areas in accordance with the division pattern information included in the management information acquired by the management information acquiring unit 106, and the feature amount acquiring unit 108 acquires from each of the areas. The feature amount identified by the feature amount identifier indicated by the management information corresponding to each area is acquired.

(Example 4)
FIG. 9 is an example of the signal processing circuit 2 in which the signal processing device 1 is configured by a circuit. The signal processing circuit 2 receives a video signal, divides two or more image signals of the video signal into two or more regions, acquires two or more feature amounts from the regions, and is determined in advance. For each period, the acquired feature values are integrated, and the integrated feature values are output to another circuit or the like.

  In FIG. 9, the signal processing circuit 2 includes a control circuit 201, a frame memory control circuit 202, an area division signal generation circuit 203, a feature amount acquisition circuit 204, a feature amount storage circuit 205, and a feature amount integration circuit 206. The feature amount storage circuit 205 includes two or more feature amount storage areas 2051. In addition, the signal processing circuit 2 usually includes a storage circuit such as a memory (not shown).

  The frame memory control circuit 202 corresponds to the input signal acquisition unit 104. The area division signal generation circuit 203 corresponds to the area division unit 107. The feature amount acquisition circuit 204 and the feature amount storage circuit 205 correspond to the feature amount acquisition unit 108 and the feature amount compression unit 110. The feature amount integration circuit 206 corresponds to the feature amount integration unit 109 and the output unit 111. A storage circuit (not shown) corresponds to the management information storage unit 101.

  The control circuit 201 receives a synchronization signal, generates a trigger for starting frame processing (hereinafter referred to as a frame processing start trigger) once per frame, and acquires a feature amount in response to the synchronization signal. Update the area setting data. The setting data is usually management information, and updating the setting data is usually acquiring one piece of management information. The management information includes a combination of a horizontal / vertical coordinate at the start of an area, a horizontal / vertical coordinate at the end of the area, and an ID number of a feature amount acquired in the area.

  An image signal is input to the frame memory control circuit 202. The frame memory control circuit 202 writes the input image signal in a storage circuit (frame memory) that stores the image signal. Also, a frame processing start trigger is received from the control circuit 201, and the latest frame for which writing has been completed and the previous frame are simultaneously read in the raster scan order. The frame memory control circuit 202 adds a frame start trigger and a line start trigger to the read image and outputs one pixel per clock.

  The area division signal generation circuit 203 receives a frame processing start trigger from the control circuit 201 and area setting data for acquiring each feature amount. When the trigger and the setting data are input, the area division signal generation circuit 203 updates the held setting data with the input setting data. In addition, a frame start trigger and a line start trigger from the frame memory control circuit 202 are input to the area division signal generation circuit 203. When the two triggers are input, the area division signal generation circuit 203 generates horizontal coordinates and vertical coordinates. When the generated coordinates coincide with the start coordinates of any feature quantity in the setting data, the region division signal generation circuit 203 outputs a feature quantity acquisition start trigger and the feature quantity ID. When the generated coordinates are within a rectangular area indicated by the start coordinates and end coordinates of any feature quantity in the setting data, the area division signal generation circuit 203 outputs a feature quantity acquisition valid signal. When the generated coordinates coincide with the end coordinates of any feature quantity in the setting data, the region division signal generation circuit 203 outputs a feature quantity acquisition end trigger.

  The feature amount acquisition circuit 204 receives the feature amount ID, the feature amount acquisition start trigger, the feature amount acquisition valid signal, and the feature amount acquisition end trigger from the region division signal generation circuit 203. When the feature amount acquisition start trigger and the feature amount ID are input, the feature amount acquisition circuit 204 initializes the circuit and switches the circuit so that the feature amount according to the feature amount ID can be acquired. Usually, the feature amount ID is used for switching the circuit. Here, it is assumed that the feature amount acquisition circuit 204 is switched so that pixel values can be acquired, for example, in order to acquire a pixel histogram. Then, the feature amount acquisition circuit 204 updates the pixel histogram based on the received data only while the feature amount acquisition valid signal is input. When a feature amount acquisition end trigger is input, the feature amount acquisition circuit 204 outputs the acquired feature amount, feature amount ID, and feature amount output trigger. Note that the feature amount acquisition circuit 204 may perform a process of reducing the data amount before outputting the feature amount. For example, when the feature amount is an inter-frame pixel difference histogram, the data amount can be reduced by processing the inter-frame pixel difference histogram into a proportion of samples whose pixel difference value is equal to or less than a certain threshold value.

  The feature quantity storage circuit 205 receives the feature quantity, the feature quantity ID, and the feature quantity output trigger from the feature quantity acquisition circuit 204. When the trigger is input, the feature amount storage circuit 205 stores the feature amount. The area in the circuit where the feature amount is stored is secured so that several frames can be stored for each feature amount ID, and the oldest data is overwritten.

  A frame processing start trigger from the control circuit 201 is input to the feature amount integration circuit 206. When the trigger is input, the feature amount integration circuit 206 reads the feature amount from the feature amount storage circuit 205 and integrates feature amounts for the past several frames for each feature amount. Then, the feature amount integration circuit 206 outputs the integrated feature amount.

(Example 5)
FIG. 10 is an example of the signal processing circuit 3 in which the signal processing device 1 is configured by a circuit. The signal processing circuit 3 acquires two or more feature amounts from the image signal and integrates the acquired feature amounts in the same manner as the signal processing circuit 2 of FIG. 9, but the feature amount is determined according to the integrated feature amount. The area for acquiring the image and the feature quantity to be acquired are changed.

  10, the signal processing circuit 3 includes a determination circuit 307 in addition to the signal processing circuit 2 of FIG. The determination circuit 307 corresponds to the determination unit 105 and the management information acquisition unit 106. Circuits other than the determination circuit 307 are the same as those in Example 4, and thus description thereof is omitted.

  The feature amount from the feature amount integration circuit 206 is input to the determination circuit 307. When the feature amount is input, the determination circuit 307 determines whether it is necessary to re-acquire the feature amount. For this determination, a feature amount condition is usually used. If the determination circuit 307 determines that it is necessary to re-acquire, the determination circuit 307 updates the setting data of the area for acquiring the feature amount according to the feature amount. Updating the setting data usually means obtaining management information corresponding to the feature amount condition.

  The area division signal generation circuit 203 receives a frame processing start trigger from the control circuit 201 and area setting data for acquiring each feature amount. In addition, the region setting signal generation circuit 203 receives region setting data for acquiring each feature amount from the determination circuit 307. As described above, when two pieces of setting data of the area for acquiring each feature amount are input, the area division signal generation circuit 203 determines the held setting data as the determination circuit 307 according to the frame processing start trigger. It is updated with the setting data of the area from which each feature amount is acquired. Since the subsequent processing is the same as that of Example 4, the description thereof is omitted.

  As described above, according to the signal processing device 1 according to the present embodiment, each unit signal is obtained by dividing two or more unit signals included in the input signal into two or more regions and acquiring different feature amounts from the regions. Thus, two or more different feature quantities can be acquired.

  Further, according to the signal processing device 1 according to the present embodiment, the accuracy of the feature amount can be improved by integrating the acquired feature amounts.

  Further, according to the signal processing device 1 according to the present embodiment, the division pattern and the feature amount acquired from each divided area can be switched for each unit signal.

  Further, according to the signal processing device 1 according to the present embodiment, the division pattern and the feature amount acquired from each divided region can be changed according to the acquired feature amount.

  In the present embodiment, the signal processing apparatus is usually determined in advance using the feature value output by the output unit 111 with respect to the input signal acquired by the input signal acquisition unit 104 as shown in FIG. The signal processing device 4 includes a signal processing unit 112 that performs signal processing and a signal processing result output unit 113 that outputs a processing result that is a result of the signal processing performed by the signal processing unit 112. The signal processing includes, for example, higher resolution and brightness adjustment. That is, the signal processing device 4 divides the input signal into two or more regions, acquires two or more different feature amounts from the two or more regions, and uses the two or more different feature amounts to input the input signal. Is a signal processing device that performs predetermined signal processing.

  Further, in each of the above embodiments, it goes without saying that two or more communication units existing in one apparatus may be physically realized by one medium.

  In each of the above embodiments, each process or each function may be realized by centralized processing by a single device or a single system, or distributed by a plurality of devices or a plurality of systems. It may be realized by being processed.

  In each of the above embodiments, each component may be configured by dedicated hardware, or a component that can be realized by software may be realized by executing a program. For example, each component can be realized by a program execution unit such as a CPU reading and executing a software program recorded on a recording medium such as a hard disk or a semiconductor memory.

  Moreover, the software which implement | achieves the signal processing apparatus in each said embodiment is the following programs, for example. That is, this program causes the computer to acquire an input signal acquisition unit that acquires two or more unit signals that are signals per unit amount determined in advance, and two or more unit signals that the input signal has. An area dividing unit that divides the image into two or more regions and acquires the two or more regions, a feature amount acquisition unit that acquires two or more different types of feature amounts from the two or more regions, and the feature amount It is a program for functioning as an output unit that outputs the feature amount acquired by the acquisition unit.

  Note that the program does not include at least processing that is performed only by hardware.

  The program may be executed by being downloaded from a server or the like, or a program recorded on a predetermined recording medium (for example, an optical disk such as a CD-ROM, a magnetic disk, a semiconductor memory, or the like) is read out. May be executed. Further, this program may be used as a program constituting a program product.

  Moreover, the computer which performs the said program may be single, and plural may be sufficient as it. That is, centralized processing may be performed, or distributed processing may be performed.

  FIG. 12 is a schematic diagram of a computer system 9 that executes the above-described program to realize the signal processing apparatus according to the above-described embodiment. The above-described embodiments can be realized by computer hardware and a computer program executed thereon.

  In FIG. 12, the computer system 9 includes a computer 901 including a CD-ROM (Compact Disk Read Only Memory) drive 9011, an FD (Flexible Disk) drive 9012, a keyboard 902, a mouse 903, and a monitor 904.

  FIG. 13 is a block diagram of the computer system 9. In FIG. 13, in addition to a CD-ROM drive 9011 and an FD drive 9012, a computer 901 includes an MPU (Micro Processing Unit) 9013 and a ROM (Read-Only Memory) 9014 for storing a program such as a bootup program. A RAM (Random Access Memory) 9015 connected to the MPU 9013 for temporarily storing instructions of the application program and providing a temporary storage space; and a hard disk 9016 for storing the application program, the system program, and data , A CD-ROM drive 9011, an FD drive 9012, an MPU 9013, and the like. Although not shown here, the computer 901 may further include a network card that provides connection to a LAN.

  A program for causing the computer system 9 to execute the functions of the signal processing apparatus and the like of the above-described embodiment is stored in the CD-ROM 9101 or the FD 9102, inserted into the CD-ROM drive 9011 or the FD drive 9012, and further the hard disk 9016. May be forwarded to. Alternatively, the program may be transmitted to the computer 901 via a network (not shown) and stored in the hard disk 9016. The program is loaded into the RAM 9015 when executed. The program may be loaded directly from the CD-ROM 9101, the FD 9102, or the network.

  The program does not necessarily include an operating system (OS) or a third-party program that causes the computer 901 to execute the functions of the signal processing apparatus according to the above-described embodiment. The program only needs to include an instruction portion that calls an appropriate function (module) in a controlled manner and obtains a desired result. How the computer system 9 operates is well known and will not be described in detail.

  The present invention is not limited to the above-described embodiments, and various modifications are possible, and it goes without saying that these are also included in the scope of the present invention.

  As described above, the signal processing device according to the present invention can obtain two or more different feature amounts from two or more unit signals of one input signal without increasing the circuit scale. And is useful as a DSP (Digital Signal Processor) or the like.

DESCRIPTION OF SYMBOLS 1 Signal processing apparatus 101 Management information storage part 102 Feature quantity identifier storage part 103 Feedback management information storage part 104 Input signal acquisition part 105 Judgment part 106 Management information acquisition part 107 Area division part 108 Feature quantity acquisition part 109 Feature quantity integration part 110 Feature Volume compression unit 111 Output unit

Claims (8)

An input signal acquisition unit that acquires an input signal having two or more unit signals that are signals per unit amount determined in advance;
An area dividing unit that divides two or more unit signals of the input signal into two or more areas;
A feature quantity acquisition unit that acquires two or more different types of feature quantities for each area from the two or more areas;
A signal processing apparatus comprising: an output unit that outputs the feature quantity acquired by the feature quantity acquisition unit. A feature quantity integrating unit that integrates two or more feature quantities of the same type acquired by the feature quantity acquiring unit for each predetermined number of unit signals;
The output unit is
The signal processing apparatus according to claim 1, wherein the feature amount integrated by the feature amount integration unit is output. Two or more pieces of management information having division pattern information indicating a pattern for dividing a unit signal into two or more regions and two or more feature amount identifiers for identifying feature amounts acquired from the two or more regions are associated with each other. A management information storage unit to be stored;
A management information acquisition unit that acquires one or more management information from the management information storage unit;
The area dividing unit includes:
According to the division pattern information included in the management information acquired by the management information acquisition unit, two or more unit signals are divided into two or more areas,
The feature amount acquisition unit
The feature amount identifier that the management information acquired by the management information acquisition unit has and that is identified by the feature amount identifier corresponding to each of the two or more regions is acquired from each of the two or more regions. Or the signal processing apparatus of Claim 2. A feature quantity identifier storing unit for storing two or more feature quantity identifiers for identifying feature quantities to be acquired from each of the two or more areas;
The area dividing unit includes:
Dividing each unit signal of two or more into two or more predetermined areas;
The feature amount acquisition unit
Two or more predetermined feature amount identifiers are acquired from the feature amount identifier storage unit, and feature amounts identified by any of the feature amount identifiers are acquired from the two or more regions in a predetermined order. The signal processing apparatus according to claim 1 or 2. One or more management information having division pattern information indicating a pattern for dividing a unit signal into two or more areas, and two or more feature quantity identifiers for identifying feature quantities acquired from the two or more areas, in association with each other; A feedback management information storage unit that stores feedback management information having a feature amount condition associated with one or more feature amount conditions;
A determination unit that determines whether the feature amount acquired by the feature amount acquisition unit or the feature amount integrated by the feature amount integration unit matches any one of the feature amount conditions included in the one or more feedback management information; ,
One or more corresponding to the feature amount condition that the feature amount acquired by the feature amount acquisition unit from the feedback management information storage unit or the feature amount integrated by the feature amount integration unit matches when the determination unit determines that they match A management information acquisition unit for acquiring the management information of
The area dividing unit includes:
According to the division pattern information included in the management information acquired by the management information acquisition unit, two or more unit signals are divided into two or more areas,
The feature amount acquisition unit
4. The feature amount identifier of the management information acquired by the management information acquisition unit and identified by the feature amount identifier corresponding to each of the two or more regions is acquired from each of the two or more regions. Or the signal processing apparatus of Claim 4. A feature amount compression unit that compresses the feature amount acquired by the feature amount acquisition unit or the feature amount integrated by the feature amount integration unit;
The output unit is
The signal processing device according to claim 1, wherein the feature amount compressed by the feature amount compression unit is output. A signal processing method performed using an input signal acquisition unit, a region division unit, a feature amount acquisition unit, and an output unit,
The input signal acquisition unit
An input signal acquisition step of acquiring an input signal having two or more unit signals which are signals per unit amount determined in advance;
The region dividing unit is
An area dividing step of dividing each of the two or more unit signals of the input signal into two or more areas;
The feature amount acquisition step includes:
A feature quantity acquisition step of acquiring two or more different types of feature quantities for each area from the two or more areas;
The output unit is
An output step of outputting the feature quantity acquired by the feature quantity acquisition unit. Computer
An input signal acquisition unit that acquires an input signal having two or more unit signals that are signals per unit amount determined in advance;
An area dividing unit that divides two or more unit signals of the input signal into two or more areas;
A feature quantity acquisition unit that acquires two or more different types of feature quantities for each area from the two or more areas;
A program for functioning as an output unit that outputs a feature amount acquired by the feature amount acquisition unit.

Images