JP2005080170A - Information processing apparatus and method, program, and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To allow a user to speedily obtain a desired processing result with simple configuration. <P>SOLUTION: When a parameter is operated via a user I/F 43 by a user 2, a feature amount extraction section 61 extracts a feature amount just before the parameter operation and a feature amount just after the operation from an internal state holding section 41. On the basis of the extracted feature amounts, a matricial equation is generated and its matrix elements are stored in a log storage section 62. When parameter prediction start is commanded via the user I/F 43 by the user 2, the feature amount extraction section 61 extracts a feature amount just before the command and supplies it to a predictive value operation unit 63. On the basis of the matrix elements stored in the log storage section 62, the predictive value operation section 63 solves the matricial equation to obtain a coefficient required for operating a predictive value and multiplies the feature amount just before the command with the coefficient to operate the predictive value. The present invention is applicable to a television receiver. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an information processing apparatus and method, a program, and a recording medium, and particularly to an information processing apparatus and method, a program, and a recording medium that allow a user to quickly obtain a desired processing result with a simple configuration. About.

  Conventionally, parameters for determining image characteristics of a television display device are set by the user himself / herself, and a satisfactory image is displayed. FIG. 1 shows the flow of conventional parameter setting processing.

  In step S1, the user operates parameters that determine the characteristics of the image to display the image. In step S2, the user determines whether or not the processing result (displayed image) is satisfied. If the user is not satisfied with the processing result, the process returns to step S1, and the parameters are operated again to display the image. If it is determined in step S2 that the user is satisfied with the processing result, the processing is terminated. In this way, the parameter operation is repeated until the user is satisfied with the processing result.

  Further, the applicant has proposed a technique for automatically setting parameters according to the history of recording the status of the selected channel so that a program viewed frequently can be automatically viewed (for example, Patent Document 1).

JP-A-8-180504

However, the conventional technique has a problem in that it is necessary to perform parameter operations many times before obtaining a processing result that satisfies the user, which places a burden on the user.
In addition, the technique disclosed in Patent Document 1 has a problem that it does not consider the setting of parameters for adjusting image characteristics and the like.

  The present invention has been made in view of such a situation, and enables a user to quickly obtain a desired processing result with a simple configuration.

  An information processing apparatus according to the present invention is an information processing apparatus that processes input data and outputs a processing result. A designation unit that designates a parameter based on a user operation, and a parameter is designated by the designation unit. If performed, an extraction means for extracting the feature quantity of the input data, an accumulation means for accumulating log data generated based on the feature quantity extracted by the extraction means, and log data accumulated in the accumulation means A prediction calculation unit that calculates a predicted value of the feature amount based on the feature amount extracted by the extraction unit, a determination unit that determines a parameter based on the prediction value calculated by the prediction calculation unit, and a determination unit And a data processing means for processing the input data based on the parameters determined by the above, and an output means for outputting the processing result by the data processing means. That.

  The extraction unit includes a first determination unit that determines whether or not parameter specification is started by the specification unit, and a second determination unit that determines whether or not the parameter specification is ended by the specification unit. Can be.

  When the first determining unit determines that the parameter specification has started, the extracting unit acquires a plurality of feature amounts of the input data immediately before the extracting unit, and the second determining unit finishes the parameter specification. If it is determined that a plurality of feature amounts of the output data immediately after that is acquired, an average value of the acquired feature amounts can be calculated.

  The extraction unit generates a determinant representing a relationship between a plurality of feature amounts of the input data and an average value of the plurality of feature amounts of the output data, and the storage unit stores the matrix elements of the determinant as log data. Can be.

  The storage means includes an initialization determination means for determining whether or not to initialize log data. If the initialization determination means determines that the log data is to be initialized, the matrix elements are stored as they are and initialized. If the determination means determines that the log data is not initialized, the matrix element can be added to the already accumulated matrix elements.

  The prediction calculation unit can generate a matrix equation based on the log data stored in the storage unit, and generate a coefficient necessary for calculating a prediction value by solving the matrix equation. .

  The designation unit receives a command for calculating a predicted value, and the extraction unit extracts a plurality of feature amounts of the input data immediately after the calculation command is received by the designation unit, and a plurality of features of the extracted input data The quantity can be supplied to the prediction calculation means.

  The prediction calculation means can calculate a prediction value based on a plurality of feature amounts and coefficients supplied from the extraction means.

  An information processing method of the present invention is an information processing method of an information processing apparatus that processes input data and outputs a processing result, and a designation determination step for determining whether or not a parameter is designated based on a user operation; When it is determined that the parameter is specified by the processing of the specification determination step, the extraction step for extracting the feature amount of the input data, and the log data generated based on the feature amount extracted by the processing of the extraction step An accumulation step for accumulating, a log data accumulated by the process of the accumulation step, a prediction computation step for computing a predicted value of the feature quantity based on the feature quantity extracted by the process of the extraction step, and a prediction computation step A determination step for determining a parameter based on the predicted value calculated by the processing, and a parameter determined by the processing of the determination step; Based on the meter, characterized in that it comprises a data processing step for processing and outputting input data.

  The program of the present invention is a program for an information processing apparatus that processes input data and outputs a processing result, and a designation determination control step for controlling determination of whether or not a parameter is specified based on a user operation; When it is determined that the parameter has been specified by the processing of the specification determination control step, the control is performed based on the extraction control step for controlling the extraction of the feature amount of the input data and the feature amount extracted by the processing of the extraction control step. Based on the accumulation control step for controlling to accumulate the log data to be accumulated, the log data accumulated by the process of the accumulation control step, and the feature quantity extracted by the process of the extraction control step. Based on the prediction calculation control step for controlling the calculation and the prediction value calculated by the processing of the prediction calculation control step, parameters are set. And a data processing control step for controlling the input data to be processed and output based on the parameters determined by the processing of the determination control step. To do.

  The recording medium of the present invention is a recording medium on which a program of an information processing apparatus that processes input data and outputs a processing result is recorded, and determines whether or not a parameter is specified based on a user operation. When it is determined that a parameter is specified by the designation determination control step to be controlled and the processing of the designation determination control step, the extraction is performed by the extraction control step for controlling the extraction of the feature amount of the input data and the processing of the extraction control step. Based on the accumulation control step for controlling to accumulate the log data generated based on the feature amount obtained, the log data accumulated by the processing of the accumulation control step, and the feature amount extracted by the processing of the extraction control step A prediction calculation control step for controlling the calculation of the predicted value of the feature amount, and a prediction calculated by the processing of the prediction calculation control step. The computer executes a determination control step for controlling the determination of the parameter based on the parameters, and a data processing control step for controlling the input data to be processed and output based on the parameter determined by the processing of the determination control step. A program to be recorded is recorded.

  In the information processing apparatus, method, and program of the present invention, a parameter is specified based on a user operation, a feature amount of input data is extracted, and a log generated based on the extracted feature amount Data is accumulated. Further, a predicted value of the feature value is calculated based on the accumulated log data and the extracted feature value, a parameter is determined based on the calculated predicted value, and an input is performed based on the determined parameter. Data is processed and processing results are output.

  According to the present invention, a user can quickly obtain a desired processing result. In particular, the user can quickly obtain a desired processing result with a simple configuration.

  Embodiments of the present invention will be described below. The correspondence relationship between the invention described in this specification and the embodiments of the invention is exemplified as follows. This description is intended to confirm that the embodiments supporting the invention described in this specification are described in the specification. Therefore, even if there is an embodiment which is described in the specification but is not described here, this means that the embodiment does not correspond to the invention. It is not a thing. Conversely, even if an embodiment is described herein as corresponding to an invention, that means that the embodiment does not correspond to an invention other than the invention. Absent.

  Further, this description does not mean that all the inventions described in the specification are claimed. In other words, this description is for the invention described in the specification and not claimed in this application, i.e., for the invention that will be filed in division or applied or added in the future. It does not deny existence.

  According to the present invention, an information processing apparatus is provided. The information processing apparatus is an information processing apparatus that processes input data and outputs a processing result, and a designation unit that designates a parameter based on a user operation (for example, user I / F 43 in FIG. 2). When the parameter is designated by the designation means, the extraction means (for example, the feature quantity extraction unit 61 in FIG. 2) that extracts the feature quantity of the input data, and the feature quantity extracted by the extraction means Based on the storage means for storing the log data generated based on the log data (for example, the log data storage unit 62 in FIG. 2), the log data stored in the storage means, and the feature amount extracted by the extraction means A prediction calculation means (for example, a prediction value calculation unit 63 in FIG. 2) for calculating the predicted value of the feature quantity, and a determination means (for example, determining a parameter based on the prediction value calculated by the prediction calculation means) For example, the parameter determination unit 64 in FIG. 2, a data processing unit for processing input data based on the parameters determined by the determination unit (for example, the display data calculation unit 42 in FIG. 2), and the data processing unit Output means (for example, the display unit 44 in FIG. 2) for outputting the processing result of the above.

  In this information processing apparatus, the extraction unit is a first determination unit that determines whether or not parameter specification is started by the specification unit (for example, the feature amount of FIG. 2 for executing the process of step S51 of FIG. 5). An extraction unit 61), and a second determination unit that determines whether or not the parameter has been specified by the specification unit (for example, the feature amount extraction unit 61 in FIG. 2 that executes the process of step S53 in FIG. 5). Can be provided.

  In the information processing apparatus, when the extraction unit determines that the parameter designation is started by the first determination unit, a plurality of feature amounts (for example, the feature amount CS0 in FIG. 3) of the input data immediately before the extraction unit are determined. To CSn-1), and when it is determined by the second determination means that the designation of the parameter has been completed, a plurality of feature amounts (for example, feature amounts CE0 to CEn in FIG. -1) is acquired, and an average value of the acquired feature values (for example, CEavg in FIG. 3) can be calculated.

  In this information processing apparatus, the extraction unit generates a determinant (for example, Expression (15)) representing a relationship between a plurality of feature amounts of the input data and an average value of the plurality of feature amounts of the output data. The storage means can store the matrix elements of the determinant (for example, the matrix elements 111 and 112 in FIG. 9) as the log data.

  The information processing apparatus includes an initialization determination unit that determines whether or not the storage unit initializes the log data (for example, the log storage unit 62 in FIG. 2 that executes step S71 in FIG. 6). And when the initialization determination unit determines to initialize the log data, the matrix elements are stored as they are, and when the initialization determination unit determines not to initialize the log data, the matrix The elements can be added to the already accumulated matrix elements (for example, the process of generating the matrix elements 91-3 and 92-3 in FIG. 8).

  In the information processing apparatus, the prediction calculation unit generates a matrix equation (for example, the matrix equation of FIG. 9) based on the log data stored in the storage unit, and solves the matrix equation, thereby predicting Coefficients necessary for calculating the values (for example, coefficients w0 to wn-1 in FIG. 9) can be generated.

  In this information processing apparatus, the specifying unit accepts a command for calculating a predicted value, and the extracting unit is configured to include a plurality of feature amounts (for example, a diagram) of input data immediately after the calculation command is received by the specifying unit. 4 feature amounts CS0 to CSn-1) can be extracted, and a plurality of feature amounts of the extracted input data can be supplied to the prediction calculation means.

  In the information processing apparatus, the prediction calculation unit calculates a prediction value (for example, calculation according to the equation (1)) based on the plurality of feature amounts supplied from the extraction unit and the coefficient. Can do.

  According to the present invention, an information processing method is provided. This information processing method is an information processing method of an information processing apparatus that processes input data and outputs a processing result, and a determination step (for example, determining whether a parameter is specified based on a user operation) When it is determined that the parameter has been designated by the processing of the designation determination step (step S51 or step S53 in FIG. 5), an extraction step (for example, step S52 in FIG. Step S54), an accumulation step (for example, step S56 in FIG. 5) for accumulating log data generated based on the feature amount extracted by the process of the extraction step, and a log accumulated by the process of the accumulation step Based on the data and the feature amount extracted by the processing of the extraction step, the predicted value of the feature amount is calculated. A measurement calculation step (for example, step S94 in FIG. 10), a determination step (for example, step S95 in FIG. 10) for determining a parameter based on the predicted value calculated by the processing in the prediction calculation step, and the determination step And a data processing step (for example, step S97 in FIG. 10) for processing and outputting the input data based on the parameters determined by the above processing.

  A program is provided by the present invention. This program is a program for an information processing apparatus that processes input data and outputs a processing result, and a specification determination control step (for example, FIG. 5) that controls determination of whether or not a parameter is specified based on a user operation. 5 (step S51 or step S53), and when it is determined that the parameter has been designated by the processing of the designation judgment control step, the extraction control step (for example, FIG. 5) controls the extraction of the feature amount of the input data. Step S52 or step S54), an accumulation control step (for example, step S56 in FIG. 5) for controlling to accumulate log data generated based on the feature amount extracted by the processing of the extraction control step, Log data accumulated by the process of the accumulation control step and extracted by the process of the extraction control step. Based on the prediction calculation control step (for example, step S94 in FIG. 10) for controlling the calculation of the prediction value of the feature amount based on the feature amount, and the prediction value calculated by the processing of the prediction calculation control step, A determination control step (for example, step S95 in FIG. 10) for controlling the determination of the parameter, and data processing control for controlling the input data to be processed and output based on the parameter determined by the processing of the determination control step. Step (for example, step S97 in FIG. 10) is executed by the computer.

  According to the present invention, a recording medium is provided. This recording medium is a recording medium in which a program of an information processing apparatus that processes input data and outputs a processing result is recorded, and controls whether or not a parameter is specified based on a user operation. When it is determined that a parameter has been specified by the designation determination control step (for example, step S51 or step S53 in FIG. 5) and the processing of the designation determination control step, the extraction of the feature amount of the input data is controlled. An extraction control step (for example, step S52 or step S54 in FIG. 5) and an accumulation control step (for example, control for accumulating log data generated based on the feature amount extracted by the processing of the extraction control step) Step S56) of FIG. 5, the log data accumulated by the process of the accumulation control step, and the extraction control A prediction calculation control step (for example, step S94 in FIG. 10) for controlling the calculation of the predicted value of the feature amount based on the feature amount extracted by the processing of the step, and the prediction calculated by the processing of the prediction calculation control step. A decision control step (for example, step S95 in FIG. 10) that controls the decision of the parameter based on the value, and the input data is processed and output based on the parameter determined by the process of the decision control step. A program for causing a computer to execute a data processing control step to be controlled (for example, step S97 in FIG. 10) is recorded.

  Embodiments of the present invention will be described below with reference to the drawings. FIG. 2 shows the overall configuration of the data prediction processing apparatus 1 to which the present invention is applied. In this example, the data prediction processing apparatus 1 configured by, for example, a television receiver mainly receives a parameter operation of the user 2, processes an image or the like, and mainly displays the processing unit 21 and displays it to the user 2. The storage unit 22 is configured to record log data based on a processing result of the processing unit 21 and predict a parameter value corresponding to an image to be displayed based on a user instruction.

  In the processing unit 21, the internal state holding unit 41 acquires the input image data, and temporarily stores (holds) it together with parameter operation information by the user 2. The image data acquired by the internal state holding unit 41 is output to the display data calculation unit 42, which displays the image data based on the parameters specified by the user I / F (Interface) unit 43. Processing (calculation) is performed, and the processing result is output to the display unit 44 and the internal state holding unit 41 as output data. The display unit 44 is configured by a CRT (Cathode Ray Tube), an LCD (Liquid Crystal Display), or the like, and displays (presents) an image to the user 2 based on output data from the display data calculation unit 42.

  The user I / F unit 43 is an interface for accepting parameter operation (designation) by the user 2, and may be configured by a switch or a button, or may be configured by a mouse, a keyboard, or the like. Good. Alternatively, a GUI (Graphical User Interface) set in advance may be displayed on the display unit 44, and the user 2 may operate a remote commander (remote controller) (not shown) to accept designation based on the GUI. .

  In the storage unit 22, the feature amount extraction unit 61 extracts a feature amount (for example, brightness) of image data at a predetermined timing from the internal state holding unit 41, and a matrix described later based on the extracted feature amount. The element is calculated and output to the log storage unit 62. The log storage unit 62 has a built-in memory such as an EEPROM (Electrically Erasable and Programmable Read Only Memory), and stores the matrix elements supplied from the feature amount extraction unit 61 in the memory as log data.

  The feature quantity extraction unit 61 also outputs the extracted feature quantity to the predicted value calculation unit 63. Based on the log data stored in the log storage unit 62 and the matrix elements supplied from the feature amount extraction unit 61, the predicted value calculation unit 63 predicts and predicts the feature amount of the image desired by the user 2. The feature amount is output to the parameter determination unit 64. The parameter determination unit 64 determines a parameter value corresponding to the feature amount based on the feature amount supplied from the predicted value calculation unit 63, and outputs the determined parameter value to the display data calculation unit 42.

  When the parameter 2 is instructed by the user 2, the display data calculation unit 42 processes the image data based on the parameter supplied from the parameter determination unit 64 and outputs the processed image data to the display unit 44. When parameter prediction is not instructed, the image data is processed based on the parameters supplied from the user I / F unit 43 and output to the display unit 44.

  Next, a method for predicting image feature amounts will be described. FIG. 3 is a diagram illustrating the operation of the parameter P by the user and the change in the image feature amount C (for example, the brightness of the image) over time, with time as the horizontal axis. Note that the value of the feature amount C changes corresponding to the operation of the parameter P.

  In this example, it is assumed that at time S, the user starts operating the parameter P, and at time E, the user finishes operating the parameter P. The time when the parameter P is operated by the user is indicated by a thick line in the figure. In this example, the user operates the parameter P for the time T1 from the time S, and then operates the parameter P for the time T2 after a predetermined time has elapsed, and then after a predetermined time interval, The parameter P is sequentially operated for T3 and T4, and the parameter P is operated a total of four times. Then, it is assumed that after operating the parameter P for the time T4 (after the fourth operation of the parameter P), an image processing result that satisfies the user is obtained.

  Now, the feature values of n images immediately before time S (parameter operation start time) are CS0, CS1,... CSn-1, respectively, and n values immediately after time E (parameter operation end time). The values of the feature values of the images are CE0, CE1,... CEn-1, and the average value of n values of CE0 to CEn-1 is CEavg. In the present invention, the average value CEavg is predicted using the values CS0 to CSn-1. Here, in order to perform prediction, a plurality of (for example, N) samples are acquired in advance, and the i-th sample Si is represented as follows. In the following expression, n Si0 to Sin-1 on the right side correspond to Cs0 to Csn-1, and one SiTRUE corresponds to CEavg.

  Si = (si0, si1, ..., sin-1, siTRUE)

  That is, the values CS0 to CSn-1 and the average value CEavg corresponding thereto can be expressed as one sample Sx as follows.

  Sx = (CS0, CS1,... CSn-1, CEavg)

  In this way, by accumulating samples and obtaining a linear prediction expression as shown in Expression (1), the predicted value ce of the feature quantity of the image is predicted. Note that c0, c1,..., Cn-1 correspond to CS0, CS1,.

  In the equation (1), the predicted value is obtained by multiplying the input value group (c0, c1,... Cn-1) by the coefficients w0, w1,. In the present invention, the coefficients w0, w1,... Wn-1 are obtained by the method of least squares.

  Now, the feature quantity of the image immediately after the parameter operation end time corresponding to the input value group (c0, c1,..., Cn-1) that is the value of the feature quantity of the n images immediately before the parameter operation start time. When an actual value (hereinafter referred to as a true value) of an average value of (n) values is cTRUE, an error between the true value cTRUE and the predicted value ce is expressed by Expression (2) and Expression (3). Can do.

  In equation (2), sample values Si = (si0, si1,..., Sin-1, siTRUE) are substituted into variables cTRUE, c0 to cn-1, and the error between the true value and the predicted value is expressed as ei. Then, the error ei can be expressed by equations (4) and (5).

  Assuming that EN is the sum of the squares of the error ei for all samples (the sum of squares), the sum of squares EN can be expressed by equations (6) to (8).

  Here, if the coefficients w0, w1,... Wn−1 can be obtained so as to minimize the square sum EN of the error ei, a predicted value close to the true value can be obtained. Therefore, when the result of partial differentiation of equation (6) with the coefficient wt is 0, a predicted value close to the true value is obtained. Therefore, equation (6) is partially differentiated, and equations (9) to (13) are can get.

の中の部分について、変形すると、次のようになる。 Of formula (13)

When the part inside is transformed, it becomes as follows.

  When this is described by a determinant, it becomes as shown in Equation (14).

  Here, since the parameter t is a parameter having values of 0, 1,..., N−1, a determinant of Expression (15) can be obtained from Expression (14).

がかかれば、式（１３）と等しくなる。すなわち、式（１５）は第ｉ番目のサンプルについてたてられた方程式であるが、Ｎ個のサンプル全てについて式（１５）と同様に方程式をたてて、

に示されるように、足しこんでいくと、式（１６）が得られる。 In equation (15)

If it is applied, it becomes equal to the equation (13). That is, equation (15) is an equation established for the i-th sample, but for all N samples, an equation is established in the same manner as equation (15),

As shown in the equation (16), equation (16) is obtained by adding the values.

  The target coefficients w0, w1,... Wn−1 are obtained by solving the matrix equation shown in Expression (16).

  In the present invention, the matrix elements of the equation (16) are accumulated as log data in the log accumulation unit 62, and the predicted value calculation unit 63 solves the matrix equation and calculates the coefficients w0, w1,. Then, the predicted value ce is calculated. The parameter determining unit 64 determines a parameter corresponding to the predicted value ce. The parameter determination will be described later.

  By doing so, as shown in FIG. 4, the values CS0, CS1,... CSn of the feature values (n) of the image immediately before the time (time S) when parameter prediction is commanded by the user. Based on −1, the predicted value ce of the feature amount of the image that is predicted to be desired by the user 2 is instantaneously calculated based on the formula (1), and the time E (the calculation of the predicted value ends, The feature value of the image output from the time when the corresponding image is output) is set as the predicted value ce.

  In this case, the operation by the user 2 is not performed on the parameter P, and the parameter value determined by the parameter determination unit 64 immediately before the time E is sufficiently short compared to the times T1 to T4 (FIG. 3). It is automatically operated and set for a certain time T10. Furthermore, the time from time S to time E shown in FIG. 4 is sufficiently shorter than the time from time S (parameter operation start time) to time E (parameter operation end time) shown in FIG. It will be a thing. As a result, an image desired by the user can be provided quickly and without imposing a burden on the user.

  Next, log data accumulation processing will be described with reference to the flowchart of FIG. This process is executed, for example, when the data prediction processing apparatus 1 is operating in an accumulation mode for accumulating data.

  In step S51, the feature amount extraction unit 61 determines whether or not a parameter operation has been performed, and waits until it is determined that a parameter operation has been performed. When the user 2 operates the user I / F 43 to change a parameter (for example, image brightness), the information (signal) is supplied to the feature amount extraction unit 61, and it is determined that the parameter has been operated. Is done.

  If it is determined in step S51 that the parameter has been operated, the process proceeds to step S52, and the feature amount extraction unit 61 acquires the feature amount CS immediately before the operation start from the internal state holding unit 41. As described above, the internal state holding unit 41 temporarily holds the contents of the image data, and holds the feature value CS of the image immediately before the start of the operation. Actually, CS is a plurality of (for example, n) values.

  In step S53, the feature amount extraction unit 61 determines whether or not the parameter operation has ended, and waits until it is determined that the parameter operation has ended. The end of the parameter operation is determined as follows, for example.

  The feature amount extraction unit 61 determines whether or not a parameter operation has been performed within a predetermined time (for example, 5 seconds) set in advance. It is determined that the operation is not completed. For example, in FIG. 3, if the scale of the time axis is 1 second, the time from the first operation (time T1) to the second operation (time T2) is 2 seconds and within 5 seconds. It is determined that the operation has not been completed yet. Similarly, the time from the second operation to the third operation (time T3) is 1 second, and the time from the third operation to the fourth operation (time T4) is 4 seconds. Yes, since both are within 5 seconds, it is determined that the operation has not been completed yet.

  Since there is no next operation within 5 seconds after the end of the fourth operation, the feature amount extraction unit 61 determines that the parameter operation has ended. Note that the end of parameter operation may be determined by another method. For example, when the user 2 inputs a parameter operation end command by operating the user I / F unit 43, it may be determined that the parameter operation has ended.

  If it is determined in step S53 that the parameter operation has ended, the process proceeds to step S54, where the feature amount extraction unit 61 acquires the feature amount CE immediately after the end of the operation from the internal state holding unit 41, and calculates the average value CEavg. calculate. In practice, CE is a plurality of (for example, n) values.

  In step S55, the feature quantity extraction unit 61 calculates the matrix elements of the determinant described above with reference to Expression (15) from CS and CEavg. In step S56, the log accumulating unit 62 executes a matrix element storing process which will be described later with reference to FIG. As a result, the matrix elements are stored in the log storage unit 62 as log data.

  Here, with reference to FIG. 6, the details of the matrix element storage processing in step S56 of FIG. 5 will be described. In step S71, the log storage unit 62 determines whether the initialization flag is ON. The initialization flag is set in advance, for example, when the user 2 operates the user I / F 43 and is stored in the memory of the accumulation unit 62. When the initialization flag is ON, the matrix element is overwritten (initialized). If it is determined in step S71 that the initialization flag is ON, the process proceeds to step S72, and the log storage unit 62 stores the matrix elements (calculated in the process of step S55) in the built-in memory. In S73, the initialization flag is set to OFF. On the other hand, if it is determined in step S71 that the initialization flag is not ON (OFF), the process proceeds to step S74, and the log storage unit 62 adds matrix elements.

  The log data accumulation process described above will be described in more detail with reference to FIGS. First, consider the case where the initialization flag is ON. As shown in FIG. 7, the feature amount extraction unit 61 acquires the feature amount CS (1) immediately before the start of the operation (step S52 in FIG. 5). In this example, five values (48, 47, 49, 46, 50) are extracted as the feature quantity CS (1). Further, the feature amount extraction unit 61 acquires the feature amount CE (1) immediately after the operation is finished, and calculates the average value CEavg (1) (step S54 in FIG. 5). In this example, five values (85, 91, 89, 88, 87) are extracted as the feature amount CE (1), and the value 88 (= (85 + 91 + 89 + 88 + 87) / 5 is calculated as the average value CEavg (1). ) Is required.

  Then, using CS (1) and CEavg (1) as one sample, a determinant similar to the above equation (15) is generated, and matrix elements 91-1 and 92-1 of the determinant are calculated. Since the initialization flag is now ON, the matrix elements 91-1 and 92-1 are stored in the log storage unit 62 as they are in step S72 of FIG. Then, the initialization flag is set to OFF (step S73).

  After that, when the user 2 operates the parameter again, as shown in FIG. 8, the feature amount extraction unit 61 causes the feature amount CS (2) immediately before the operation start and the feature amount CE (2) immediately after the operation end. A new value is obtained, and the average value CEavg (2) is newly calculated. In this example, five values of (22, 20, 24, 22, 25) are extracted as the feature quantity CS (2), and (35, 35, 36, 35, 34) as the feature quantity CE (2). ) Are extracted, and the value 35 (= (35 + 35 + 36 + 35 + 34) / 5) is obtained as the average value CEavg (2).

  Then, using CS (2) and CEavg (2) as one sample, a determinant is generated as in the case of FIG. 7, and matrix elements 91-2 and 92-2 of the determinant are calculated. Now, since the initialization flag is OFF, the matrix elements 91-2 and 92-2 are added to the matrix elements stored in the log storage unit 62 in step S74 of FIG. That is, the matrix elements 91-2 and 92-2 in FIG. 8 are added to the matrix elements 91-1 and 92-1 in FIG. 7, and the matrix elements 91-3 and 92-3 are logged in the log storage unit 62. Stored as data. As described above, as in the above-described equation (16), a determinant is generated for each sample for a plurality of samples, and addition of each determinant is performed.

  In FIGS. 7 and 8, the left-down hatching represents the feature value used to generate the accumulated data, and the right-down hatching represents the accumulated data.

  When the determinant shown in FIG. 8 is replaced with the equation (16), the matrix elements 111 and 112 shown in FIG. 9 are stored (stored) in the memory of the log storage unit 62. By doing so, it is not necessary to store the values of a plurality of samples, and the capacity of the memory of the log storage unit 62 can be reduced. For example, when the number of feature amounts extracted by the feature amount extraction unit 61 is n, the matrix elements to be stored are n × n (matrix elements 111) + n (matrix elements 112), and n (n + 1) pieces are obtained. Become.

  Even if the value of each matrix element is estimated as a long double (80 bits) value on a computer, the memory of the log storage unit 62 only needs to have a storage capacity of 80n (n + 1) bits. When the number n is 10, the memory of the log storage unit 62 can be configured with a memory having a storage capacity of 1.1 Kbytes (= 80 × 10 (10 + 1) = 8800 bits / 8 = 1100 bytes), and at low cost. , Accurate data prediction can be realized.

  In this example, when the initialization flag is OFF, the addition of the matrix element is performed. However, the initialization flag may not be provided and the addition of the matrix element may be always performed.

  In this way, each time the user 2 operates the parameter, a feature amount sample is extracted, a matrix element corresponding to the sample is generated, and the matrix element is stored in the log storage unit 62 as log data. Since matrix elements are added each time a sample is extracted, it is possible to accumulate log data that more accurately reflects the characteristics of the parameter operation of the user 2.

  Normally, the storage mode is set, and the log data storage process described with reference to the flowchart of FIG. 5 is performed. When the prediction mode is set by the user, the parameter prediction process shown in FIG. 10 is executed.

  In step S91, the feature quantity extraction unit 61 determines whether or not parameter prediction is commanded, and waits until it is determined that parameter prediction is commanded. The parameter prediction command is performed, for example, when the user 2 operates the user I / F 43. If it is determined in step S91 that parameter prediction has been commanded, the process proceeds to step S92, and the feature amount extraction unit 61 acquires the feature amount of the image immediately before the command. At this time, a plurality of values are extracted as feature amounts.

  On the other hand, if it is determined in step S91 that parameter prediction has been commanded, the information is also supplied to the predicted value calculation unit 63. In step S93, the predicted value calculation unit 63 generates a matrix equation as shown in FIG. 9 based on the matrix elements 111 and 112 stored in the log storage unit 62, solves the matrix equation, and calculates the coefficient. Calculate w0 to wn-1.

  In step S94, the predicted value calculation unit 63 multiplies the value of the feature amount acquired in step S92 by the coefficient calculated in step S93, and calculates the predicted value according to the above-described equation (1).

  In step S95, the parameter determination unit 64 calculates a parameter value corresponding to the predicted value calculated in step S94. Now, if the feature quantity of the image is represented by C and the value of the parameter is represented by P, it is assumed that the feature quantity C and the parameter P can be related by a function F. That is, it is assumed that the following equation holds between the feature amount C and the parameter P.

  C = F (P)

In step S95, P = F ⁻¹ (C) is calculated with the feature amount predicted and calculated in step S94 as C, and the parameter value P is calculated. For example,
Feature value C = parameter value P × 2 + 12
When the feature amount calculated in step S94 is 70, the following equation is established.

  70 = P × 2 + 12

  Accordingly, the parameter value P is calculated as 29 (= (70-12) / 2).

  In step S96, the parameter determination unit 64 outputs the value of the parameter calculated in step S95 to the display data calculation unit 42. In step S97, the display data calculation unit 42 processes the image data supplied from the internal state holding unit 41 based on the parameters output in step S96, outputs the processing result to the display unit 44, and displays the display unit 44. Displays an image.

  In this way, a predicted value estimated to be desired by the user 2 is calculated, and a parameter corresponding to the predicted value is automatically set. As a result, as described above, an image predicted to be desired by the user 2 can be provided in a sufficiently short time without causing the user 2 to operate the parameters.

  In this example, the example in which the parameter is predicted for the image data has been described, but the predicted data is not limited to the image data. For example, parameters may be similarly predicted for audio data.

  It does not matter whether the above-described series of processing is realized by hardware or software. When the above-described series of processing is executed by software, a program constituting the software executes various functions by installing a computer incorporated in dedicated hardware or various programs. For example, a general-purpose personal computer as shown in FIG. 11 is installed from a network or a recording medium.

  In FIG. 11, a CPU (Central Processing Unit) 901 executes various processes according to a program stored in a ROM (Read Only Memory) 902 or a program loaded from a storage unit 908 to a RAM (Random Access Memory) 903. To do. The RAM 903 also appropriately stores data necessary for the CPU 901 to execute various processes.

  The CPU 901, ROM 902, and RAM 903 are connected to each other via a bus 904. An input / output interface 905 is also connected to the bus 904.

  The input / output interface 905 includes an input unit 906 including a keyboard and a mouse, a display (display unit) including a CRT (Cathode Ray Tube) and an LCD (Liquid Crystal display), an output unit 907 including a speaker, a hard disk, and the like A storage unit 908 composed of a communication unit 909 composed of a modem and a terminal adapter is connected. The communication unit 909 performs communication processing via a network such as the Internet.

  A drive 910 is also connected to the input / output interface 905 as required. For example, a removable medium 911 is mounted on the drive 910 as a recording medium on which the program of the present invention is recorded, and read from them. A computer program is installed in the storage unit 908 as necessary.

  Note that the steps of executing the series of processes described above in this specification are performed in parallel or individually even if they are not necessarily processed in time series, as well as processes performed in time series in the order described. The processing to be performed is also included.

It is a flowchart explaining the conventional parameter setting process. It is a block diagram which shows the structural example of the data prediction processing apparatus of this invention. It is a figure which shows the relationship between the operation of a parameter, and the change of a feature-value. It is a figure which shows the relationship between the feature-value before prediction and the feature-value after prediction. It is a flowchart explaining a log data accumulation | storage process. It is a flowchart explaining a matrix element storage process. It is a figure which shows the example of the matrix element produced | generated based on the sample of a feature-value. It is a figure which shows the example in which another matrix element is added to the matrix element of FIG. It is a figure which shows the matrix element accumulate | stored as log data. It is a flowchart explaining a parameter prediction process. And FIG. 16 is a block diagram illustrating a configuration example of a personal computer.

Explanation of symbols

  1 data prediction processing device, 21 processing unit, 22 storage unit, 41 internal state holding unit, 42 display data calculation unit, 43 user I / F unit, 61 feature quantity extraction unit, 62 log storage unit, 63 predicted value calculation unit, 64 Parameter determination unit

Claims (11)

An information processing apparatus that processes input data and outputs a processing result,
A specifying means for specifying parameters based on user operations;
An extraction unit that extracts a feature amount of input data when a parameter is specified by the specifying unit;
Storage means for storing log data generated based on the feature amount extracted by the extraction means;
Prediction calculation means for calculating a predicted value of the feature quantity based on the log data accumulated in the accumulation means and the feature quantity extracted by the extraction means;
Determining means for determining a parameter based on the predicted value calculated by the prediction calculating means;
Data processing means for processing input data based on the parameters determined by the determination means;
An information processing apparatus comprising: output means for outputting a processing result by the data processing means. The extraction means includes
First determination means for determining whether or not parameter designation is started by the designation means;
The image signal processing apparatus according to claim 1, further comprising: a second determination unit that determines whether or not the parameter specification has been completed by the specification unit. The extraction unit obtains a plurality of feature amounts of the input data immediately before the parameter determination by the first determination unit, and specifies the parameter by the second determination unit. 3. The information processing apparatus according to claim 2, wherein when it is determined that a plurality of feature amounts of the output data immediately after that is acquired, an average value of the acquired feature amounts is calculated. The extraction unit generates a determinant representing a relationship between a plurality of feature amounts of the input data and an average value of the plurality of feature amounts of the output data,
The information processing apparatus according to claim 3, wherein the storage unit stores the matrix elements of the determinant as the log data. The storage means includes
An initialization determination means for determining whether to initialize the log data;
If it is determined by the initialization determining means to initialize the log data, the matrix elements are stored as they are,
The information processing apparatus according to claim 4, wherein, when the initialization determination unit determines that the log data is not initialized, the matrix element is added to the already accumulated matrix elements. The prediction calculation unit generates a matrix equation based on the log data stored in the storage unit, and generates a coefficient necessary for calculating a prediction value by solving the matrix equation. The information processing apparatus according to claim 5. The designation means accepts a command for calculating a predicted value,
The extraction means extracts a plurality of feature quantities of input data immediately after the instruction of the calculation is received by the designation means, and supplies the plurality of feature quantities of the extracted input data to the prediction calculation means. The information processing apparatus according to claim 1. The information processing apparatus according to claim 7, wherein the prediction calculation unit calculates a prediction value based on a plurality of feature amounts supplied from the extraction unit and the coefficient. An information processing method for an information processing apparatus that processes input data and outputs a processing result,
A designation determination step for determining whether or not a parameter is specified based on a user operation;
An extraction step of extracting a feature amount of input data when it is determined that the parameter is specified by the processing of the specification determination step;
An accumulation step of accumulating log data generated based on the feature amount extracted by the processing of the extraction step;
A predictive calculation step of calculating a predicted value of the feature quantity based on the log data accumulated by the process of the accumulation step and the feature quantity extracted by the process of the extraction step;
A determination step for determining a parameter based on the predicted value calculated by the process of the prediction calculation step;
And a data processing step of processing and outputting the input data based on the parameter determined by the processing of the determining step. A program of an information processing apparatus that processes input data and outputs a processing result,
A designation determination control step for controlling determination of whether or not a parameter is specified based on a user operation;
An extraction control step for controlling the extraction of the feature amount of the input data when it is determined that the parameter has been specified by the process of the designation determination control step;
An accumulation control step for controlling to accumulate log data generated based on the feature amount extracted by the processing of the extraction control step;
A prediction calculation control step for controlling calculation of a predicted value of the feature quantity based on the log data accumulated by the process of the accumulation control step and the feature quantity extracted by the process of the extraction control step;
A determination control step for controlling the determination of parameters based on the predicted value calculated by the processing of the prediction calculation control step;
A program that causes a computer to execute a data processing control step of controlling input data to be processed and output based on parameters determined by the processing of the determination control step. A recording medium on which a program of an information processing apparatus that processes input data and outputs a processing result is recorded,
A designation determination control step for controlling determination of whether or not a parameter is specified based on a user operation;
An extraction control step for controlling the extraction of the feature amount of the input data when it is determined that the parameter has been specified by the process of the designation determination control step;
An accumulation control step for controlling to accumulate log data generated based on the feature amount extracted by the processing of the extraction control step;
A prediction calculation control step for controlling calculation of a predicted value of the feature quantity based on the log data accumulated by the process of the accumulation control step and the feature quantity extracted by the process of the extraction control step;
A determination control step for controlling the determination of parameters based on the predicted value calculated by the processing of the prediction calculation control step;
A recording medium on which is recorded a program that causes a computer to execute a data processing control step of controlling input data to be processed and output based on parameters determined by the processing of the determination control step .

Images