JP5444822B2 - Genetic processing apparatus, genetic processing method and program - Google Patents

Description

  The present invention relates to a genetic processing apparatus, a genetic processing method, and a program.

  A filter string generation method using evolutionary computation such as a genetic algorithm or genetic programming is known. In this method, operations such as crossover, mutation, and selection are repeated a plurality of times for a filter row to generate a new filter row. According to such a filter string generation method using evolutionary computation, a filter string having a complex structure that is optimal for each actual application case and difficult to obtain analytically can be obtained with less effort. Can be designed.

Masayoshi Maebuchi and two others, “Study on Image Filter Design Using Genetic Algorithm” [online], Computer Utilization Education Council, [March 20, 2008 search], Internet <URL: http: //www.ciec. or.jp/event/2003/papers/pdf/E00086.pdf>

  However, in the filter row generated by such evolutionary calculation, a repeated portion in which a plurality of the same filter components are connected may appear. In this case, the calculation of the filter row is wasteful, and thus the calculation cost may be wasted.

In order to solve the above problems, in the first aspect of the present invention, a new conversion is performed by genetic processing from at least one converter that combines a plurality of processing components that process input data and output as output data. A generating unit that generates a device, and an integration unit that integrates two or more processing components linked by connection between input and output in a new converter, and replaces the processing components with processing corresponding to the two or more processing components And a dividing unit that divides one processing component of the plurality of processing components into two or more processing components, and the dividing unit is one in which two or more processing components are integrated and replaced by the integration unit Provided are a genetic processing device that divides a processing component into a set of two or more processing components, and a genetic processing method and program related to the genetic processing device .
In the second aspect of the present invention, from at least one converter that combines a plurality of processing components that process input data and output as output data, a generation unit that generates a new converter by genetic processing; An integration unit that integrates two or more processing components linked by connection between input and output in a new converter and replaces the processing components with processing components corresponding to the two or more processing components. Each of the components is a filter component that converts input data into output data, and the integration unit integrates two or more identical filter components connected by connection between input and output, and the number of filter components to be integrated Provided are a genetic processing device that replaces an integrated filter component that performs conversion with an intensity of minutes, and a genetic processing method and program related to the genetic processing device.
In the third aspect of the present invention, a generation unit that generates a new converter by genetic processing from at least one converter that combines a plurality of processing components that process input data and output as output data; An integration unit that integrates two or more processing components linked by connection between input and output in a new converter and replaces the processing components with processing components corresponding to the two or more processing components. Each of the components is a filter component that converts input data into output data, and the integration unit uses the same one filter component for conversion by the same two or more filter components connected by connection between input and output. A genetic processing device that replaces the same two or more filter components with the same one filter component when the same as the conversion, and the genetic processing device Providing heat treatment method, and a program.
In a fourth aspect of the present invention, a generating unit that generates a new converter by genetic processing from at least one converter that combines a plurality of processing components that process input data and output as output data; An integration unit that integrates two or more processing components linked by connection between input and output in a new converter and replaces the processing components with processing components corresponding to the two or more processing components. Each of the parts is a filter part that converts input data into output data, and the integration unit integrates repeated parts in which the filter part columns are repeatedly connected, and performs conversion by the filter part column for the number of repetitions. A genetic processing device that replaces at least one filter component at a strength of the same, and a genetic processing method and program related to the genetic processing device To.

  It should be noted that the above summary of the invention does not enumerate all the necessary features of the present invention. In addition, a sub-combination of these feature groups can also be an invention.

FIG. 1 shows a configuration of a genetic processing apparatus 100 according to the present embodiment. FIG. 2 shows an example of a converter having a configuration in which the processing components 22 according to this embodiment are combined in series. FIG. 3 shows an example of the converter 20 having a configuration in which the processing component 22 according to this embodiment is combined with a tree structure. FIG. 4 shows an example of a genetic operation performed on the converter 20 having a configuration in which the processing components 22 according to this embodiment are combined in series. FIG. 5 shows an example of a method for calculating similarity, which is an example of a parameter representing the adaptability of the converter according to the present embodiment. FIG. 6 shows a processing flow of the genetic processing apparatus 100 according to the present embodiment. FIG. 7 shows an example of the processing flow of the genetic processing apparatus 100 in step S640 of FIG. FIG. 8 is a diagram illustrating a converter 810 in which the same processing components according to the present embodiment are repeatedly connected and converters 820 and 830 in which the number of repetitions is changed. FIG. 9 is a diagram showing a converter 910 in which the same processing component sequence according to the present embodiment is repeatedly connected and a converter 920 in which the number of repetitions is changed. FIG. 10 shows an example of a hardware configuration of a computer 1900 according to the present embodiment.

  Hereinafter, the present invention will be described through embodiments of the invention, but the following embodiments do not limit the invention according to the claims. In addition, not all the combinations of features described in the embodiments are essential for the solving means of the invention.

  FIG. 1 shows a configuration of a genetic processing apparatus 100 according to the present embodiment. The genetic processing apparatus 100 merges a plurality of processing components in the converter during or after generating a converter suitable for converting input data to target data using evolutionary computation. Thus, the converter is simplified and the calculation cost of the converter is reduced. The genetic processing device 100 includes a converter storage unit 110, a generation unit 120, a learning data storage unit 130, a processing unit 140, a fitness calculation unit 150, a selection unit 160, an integration unit 170, a processing unit A component storage unit 180 and a dividing unit 190 are provided.

  The converter storage unit 110 stores a converter. The converter may be configured by combining a plurality of processing components that convert an input data set including one or more input data into an output data set including one or more output data. The converter may have a configuration in which processing components are combined in series. The converter may have a configuration in which processing components are combined in a tree structure or the like. Further, the converter may be a program that performs an operation on the data set, an arithmetic expression that represents the operation content to be performed on the data set, or the like.

  The data set to be converted by the converter may be a one-dimensional data string, a two-dimensional data group, a three-dimensional data group, or a further multidimensional data group. The one-dimensional data string may be, for example, time series data or an array data string. The two-dimensional data group may be image data or the like in which a plurality of pixel data and the like are arranged in a two-dimensional space. The three-dimensional data group may be volume data or the like in which data values representing color or density are arranged at each grid point in the three-dimensional space. The converter may output a data set having a dimension different from that of the input data set.

  In one example, each of the plurality of processing components is a filter component that converts an input data set into an output data set. When the input data set is image data, each processing component may be a filter component that converts an input image including a plurality of pixel data into a converted image including a plurality of pixel data. Each converter may be an image filter in which a plurality of filter components are combined.

  The generation unit 120 generates a new converter by genetic processing from at least one converter that connects and combines input and output of a plurality of processing components that process input data and output as output data. The genetic process will be described later with reference to FIGS.

  The learning data storage unit 130 stores learning data. The learning data includes learning input data and learning target data. In one example, the converter to be generated by genetic processing converts learning input data into learning target data. When an image filter is used as each converter, the learning input data may be a learning input image, and the learning target data may be a learning target image. For example, the learning input image may be an inspection target image obtained by imaging an inspection target including a defect, and the learning target image is an extraction image of the defect that is a target to be extracted from the inspection target image. Also good.

  The processing unit 140 receives a converter from the generation unit 120 and receives learning input data from the learning data storage unit 130. Subsequently, the processing unit 140 converts the learning input data using each converter to generate output data.

  The fitness calculation unit 150 receives the output data from the processing unit 140 for each converter. Subsequently, the fitness level calculation unit 150 calculates the fitness level for conversion from the learning input data to the learning target data. Here, the fitness is an index value indicating whether or not the learning input data is suitable to be converted into learning target data. In this embodiment, the higher the fitness value, the more the learning input data is represented. Indicates that it is suitable for conversion into learning target data. In one example, the fitness level calculation unit 150 calculates a similarity level of the learning input image converted by the processing unit 140, which will be described later with reference to FIG.

  The selection unit 160 selects a converter based on the degree of matching for each converter and the number of processing components. In one example, the selection unit 160 selects a converter for each converter based on at least one of the fitness, the number of processing components, and the processing load. In this way, the selection unit 160 can preferentially select a converter having a higher degree of fitness, a converter having a smaller number of processing components, and / or a converter having a smaller processing load in the present embodiment. .

  The integration unit 170 integrates two or more processing components linked by connection between input and output in a new converter, and replaces the processing components with processing corresponding to the two or more processing components. Here, the processing corresponding to the two or more processing components means that, when input data is given, output data that is the same as or similar to the output data obtained by processing the input data by the two or more processing components. It is a process to obtain. The integration unit 170 may replace two or more processing components with a processing component that performs processing that combines processing by the two or more processing components.

  In one example, the integration unit 170 integrates repetitive portions in which processing component sequences are repeatedly connected, and replaces the processing component sequence with at least one processing component that performs conversion with the intensity corresponding to the number of repetitions. Here, the column of processing components refers to a structure in which one or more processing components are connected. In this way, even when the processing component columns are repeatedly connected, the learning data storage unit 130 can reduce the number of processing components by replacing it with one processing component column that provides the same or similar effect. Therefore, the calculation cost is reduced.

  The processing component storage unit 180 stores processing components. When two or more processing components stored in the processing component storage unit 180 are connected, the integration unit 170 may replace the same two or more processing components with the same one processing component.

  The processing component storage unit 180 stores processing component identification information in which conversion by the same two or more processing components connected by connection between input and output is the same as conversion by the same one processing component. When the processing component is a filter component, the processing component storage unit 180 is a filter storage unit. In one example, when derived from a class of an object-oriented program, the processing component storage unit 180 stores the class name of the same processing component as identification information. In one example, when the processing component is provided by a program stored in a file, the processing component storage unit 180 stores the file name of the file as identification information. Thereby, the integration unit 170 can replace two or more newly used processing components with the same processing component, so that the calculation cost is reduced.

  The dividing unit 190 divides one processing component among the plurality of processing components into two or more processing components. The one processing component may be one obtained by integrating and replacing two or more processing components by the integration unit 170. When the processing component stored in the processing component storage unit 180 is included, the dividing unit 190 may divide the processing component into a set of two or more processing components.

  In one example, the dividing unit 190 divides at least one converter into two or more partial converters by dividing one processing component of at least one converter into two or more processing components. Subsequently, the generation unit 120 sets some of the two or more partial converters as at least a part of the new converter. As a result, the genetic processing apparatus 100 can reduce the number of converters in each generation and the calculation cost of each converter by integrating the processing parts, and can divide and crossover the combined processing parts again. Various genetic operations can be performed.

  FIG. 2 shows an example of the converter 20 having a configuration in which the processing components 22 according to this embodiment are combined in series. FIG. 3 shows an example of the converter 20 having a configuration in which the processing component 22 according to this embodiment is combined with a tree structure. The converter 20 receives input data, performs a filter operation process on the received input data, and outputs output data.

  The converter 20 has a configuration in which a plurality of processing components 22 are combined. The converter 20 may have a configuration in which the processing components 22 are combined in series as shown in FIG. Moreover, the converter 20 may have the structure which combined the process component 22 with the tree structure, as FIG. 3 shows.

  The converter 20 having a configuration in which the processing components 22 are combined in a tree structure receives input data from the processing components 22 at the end of the tree structure, and outputs output data from the highest processing component 22 in the tree structure. Further, in such a converter 20, the same input data is given to each of the plurality of terminal processing components 22. Alternatively, such a converter 20 may be provided with different input data for each of the plurality of terminal processing components 22.

  Each of the plurality of processing components 22 may be a program module, an arithmetic expression, or the like. The processing component 22 receives data output from the processing component 22 arranged in the preceding stage, performs an operation on the received data, and gives the processed data to the processing component 22 arranged in the subsequent stage.

  Each of the plurality of processing components 22 is a unary operation such as a binarization operation, a histogram operation, a smoothing operation, an edge detection operation, a morphology operation and / or an operation in a frequency space (for example, a low-pass filtering operation and a high-pass filtering operation). You may do. Further, each of the plurality of processing components 22 has an average operation, a difference operation, and / or a fuzzy operation (for example, an OR operation, an AND operation, an algebraic sum, an algebraic product, a limit sum, a limit product, an intense sum and an intense product). Binary operations such as

  FIG. 4 shows an example of a genetic operation performed on the converter 20 having a configuration in which the processing components 22 according to this embodiment are combined in series.

  The generation unit 120 may generate a new two or more transducers 20 by performing a crossover operation, which is an example of a genetic operation, with respect to two or more transducers 20. As illustrated in FIG. 4, the generation unit 120 processes a part of the processing component group 24 </ b> A of at least one converter 20 </ b> A that has already been generated, and processes at least a part of another converter 20 </ b> B that has already been generated. New transducers 20E and 20F may be generated by replacing the component group 24B. The processing component group 24 is a member in which one or a plurality of processing components 22 are combined.

  The generation unit 120 may generate a new converter 20 by performing a mutation operation, which is an example of a genetic operation, on the single converter 20. As illustrated in FIG. 4, the generation unit 120 replaces a part of the processing component group 24 </ b> C of the already generated one converter 20 </ b> C with another processing component group 24 </ b> G selected at random, and creates a new processing component group 24 </ b> G. The converter 20G may be generated.

  The generation unit 120 may leave the current-generation converter 20 as the next-generation converter 20 as it is. As illustrated in FIG. 4, the generation unit 120 may generate a next-generation converter 20H that includes the configuration of the processing component 22 of the converter 20D as it is.

  The selection unit 160 selects one or a plurality of converters 20 by a technique in which a natural selection of a living organism is modeled with respect to the plurality of converters 20 generated by the generation unit 120. The selection unit 160 may preferentially select the converter 20 having a higher matching degree among the plurality of converters 20. The selection unit 160 may select the converter 20 according to a technique such as elite selection and roulette selection based on the suitability of each of the plurality of converters 20. Then, the selection unit 160 stores the converter 20 in the converter storage unit 110 in order to make the selected converter 20 survive to the next generation, and the converter storage unit 110 in order to kill the converter 20 that has not been selected. Delete from within.

  FIG. 5 shows an example of a similarity calculation method that is an example of a parameter that represents the degree of fitness of the converter 20 when the converter 20 according to the present embodiment is an image filter. In the present embodiment, the fitness level calculation unit 150 calculates a similarity level indicating how similar the output image generated by converting the input image by the converter 20 and the target image. This similarity is an example of a parameter representing the degree of fitness of the converter 20 and is used as an evaluation value or index indicating that the converter 20 that generated the output image is more appropriate as the value is larger. .

  The fitness calculation unit 150 calculates a comparison value by comparing the value of each region of the output image with the value of each region corresponding to the target image, and assigns a weight specified by the weight data to the comparison value for each region. Multiply. Then, the fitness level calculation unit 150 calculates a value obtained by averaging or summing the comparison values for each area multiplied by the weight for all areas, and outputs the calculated value as the similarity of the output image.

  The fitness level calculation unit 150 may calculate a difference or a ratio between the luminance value for each pixel of the output image and the luminance value of the corresponding pixel of the target image. Then, the fitness calculation unit 150 multiplies each calculated difference or ratio for each pixel by the corresponding weight specified by the weight data, and sums or averages the difference or ratio for each pixel multiplied by the weight. The similarity may be calculated.

  For example, when the weight data is represented by a weight image, the fitness level calculation unit 150 may calculate the similarity level by performing an operation represented by the following formula (1). In this case, the sizes of the output image, the target image, and the weight image are the same.

  In the formula (1), f represents the similarity. Ymax represents the maximum luminance value.

  X is a variable indicating the pixel position in the horizontal direction of the image. y is a variable indicating the pixel position in the vertical direction of the image. xmax is a constant indicating the number of pixels in the horizontal direction of the image. ymax is a constant indicating the number of pixels in the vertical direction of the image.

  Iweight (x, y) represents the luminance value of the pixel at the (x, y) position in the weighted image. Itarget (x, y) represents the luminance value of the pixel at the (x, y) position in the target image. Ifilter (x, y) represents the luminance value of the pixel at the (x, y) position in the output image.

  That is, as shown in the numerator part in the curly braces of Equation (1), the fitness calculation unit 150 calculates the luminance value of the weighted image with respect to the absolute value of the difference between the luminance value of the target image and the luminance value of the output image. Is calculated for every pixel in the screen, and a total weighted difference value is calculated by adding the calculated weighted difference values for all pixels. Further, as shown in the denominator part in the curly braces of Expression (1), the fitness level calculation unit 150 calculates a total weight obtained by adding the luminance values of the weighted image for all pixels. Furthermore, the fitness calculation unit 150 normalizes the division value obtained by dividing the total weighted difference value by the total weight (in the curly braces of Expression (1)) by the reciprocal (1 / Ymax) of the maximum luminance value. The value (the second term in equation (1)) is calculated. Then, the fitness level calculation unit 150 calculates a value obtained by subtracting the normalized value from 1 as the similarity level f. In this way, the fitness level calculation unit 150 can calculate the similarity level by weighting the difference between the output image and the target image for each region.

  FIG. 6 shows a processing flow of the genetic processing apparatus 100 according to the present embodiment. The genetic processing device 100 repeatedly executes the processes in steps S610 to S670 a plurality of times (for example, a plurality of generations) (S610, S670).

  First, the genetic processing device 100 generates a converter 20 that is randomly generated or has already been generated as an initial converter 20 for generating a converter 20 that converts given learning input data into learning target data. Is stored in the converter storage unit 110 in advance.

  In each generation, first, the generation unit 120 performs a genetic operation such as a crossover operation and a mutation operation on the plurality of transducers 20 remaining from the previous generation to generate a plurality of new transducers 20. (S620). For example, in the first generation, the generation unit 120 may perform a genetic operation on the converter 20 stored in advance in the converter storage unit 110 to generate a plurality of new converters 20. Good.

  Subsequently, the processing unit 140 converts the input data for learning by each of the plurality of converters 20 remaining from the previous generation and the plurality of converters 20 newly generated in step S12 to generate output data ( S630). Accordingly, the processing unit 140 can generate a plurality of output data corresponding to the plurality of converters 20.

  Subsequently, the fitness level calculation unit 150 calculates the fitness level between each of the plurality of output data and the learning target data, and the selection unit 160 selects the plurality of converters 20 corresponding to the output data having a higher fitness level. (S640). Note that, in the last generation, the selection unit 160 may select one converter 20 corresponding to one output data having the highest fitness.

  Subsequently, the selection unit 160 causes the one or more converters 20 selected in step S650 to remain in the next generation, and deletes the converter 20 that generated the output image not selected in step S640 (S650).

  Subsequently, the integration unit 170 integrates the same two or more processing components 22 connected by connection between the input and output of the processing components 22 in the one or more converters 20 selected in step S650. The processing component 22 is converted with the intensity corresponding to the number of the processing components 22 to be integrated (S660). In step S660, when the processing component 22 is a filter component, for example, the integration unit 170 may replace n connected expansion / contraction filters with an n-fold expansion / contraction filter. Further, for example, when n blur filters, median filters, average filters, histogram expansion / compression filters, etc. are connected in step S660, the n filters may be replaced with filters that perform n times the intensity. Good. In this way, the integration unit 170 can reduce the number of processing components 22 by replacing the same processing component 22 with one processing component 22 that provides the same or similar effect with respect to the input data. Decrease.

  The integration unit 170 integrates the two or more processing components 22 connected by the connection between the input and output of the processing components 22 in the one or more converters 20 selected in step S650, and connects the two or more connected components. Alternatively, the processing component 22 may be replaced with a product matrix of conversion matrices representing each of the processing components 22. In this way, when the converter 20 is actually used, the integrated processing component 22 applies the conversion matrix that is a product matrix once instead of multiplying the input vector data by two or more conversion matrices. Since multiplication is performed, the calculation cost is reduced.

  When the conversion by the same two or more processing components 22 linked by the connection between the input and output is the same as the conversion by the same one processing component 22, the integration unit 170 performs the same two or more processing components. 22 may be replaced by the same processing component 22. As such a processing component 22, for example, for each region of given image data, a binarization processing component that assigns 0 to the dark region and 255 to the bright region, each numerical value of the digitized input data A processing component for assigning an absolute value to a processing component, a processing component having a power matrix as a transformation matrix, and the like. In this way, the integration unit 170 can replace the two or more processing components 22 with one processing component 22 and reduce the number of processing components 22, thereby reducing the calculation cost.

  Subsequently, the dividing unit 190 divides the one processing component 22 replaced by integrating the two or more processing components 22 by the integration unit 170 into a set of two or more processing components 22 (S665). For example, when the processing component 22 is a filter component, the dividing unit 190 divides the integrated filter component into a set of at least two filter components that perform the same conversion as the integrated filter component together. Here, the integrated filter component is a filter component replaced by the integration unit 170. For example, the median filter integrated by the addition of the number of pixels, the maximum / minimum luminance integrated by the addition of the number of pixels ( An Adoption filter, a multiplication filter synthesized by a product of multipliers, and a maximization / minimization filter synthesized by selecting the larger / smaller threshold, or a binarization filter. Further, the dividing unit 190 is the same having the intensity of a times and b times, where n = a + b, where n = a + b is an integrated filter component having a strength of n times, such as a blur filter, a median filter, an average filter, and a histogram expansion / compression filter. It may be replaced with synthesis of filter parts that perform conversion.

  Here, the dividing unit 190 may selectively divide the integrated filter component. For example, the dividing unit 190 may divide the integrated filter component every several generations, or may divide only a specific type of integrated filter component.

  When the generation unit 120 divides the processing component 22 for the purpose of causing the generation unit 120 to cross the plurality of converters 20, the generation unit 120 instructs the division unit 190 to divide the processing component 22. Also good. Accordingly, the dividing unit 190 can perform division suitable for the generating unit 120 to perform various genetic processes.

  The genetic processing device 100 repeatedly executes the processing of steps S620 to S665 for a plurality of generations (for example, several tens or hundreds of generations), and after executing the processing up to the Nth generation (N is an integer of 2 or more), The flow is exited (S670). In this way, the genetic processing device 100 can generate the converter 20 suitable for converting the learning input data into the learning target data using the evolutionary calculation.

  In one embodiment shown in FIG. 6, the integration unit 170 replaces the same two or more processing components 22 or processing components 22 in the same processing component 22 or processing component 22 between steps S650 and S665. It is replaced with the column. However, the integration unit 170 may replace the same two or more processing components 22 or processing component 22 columns with the same processing component 22 or processing component 22 sequence at any stage.

  In one example, the generation unit 120 further generates another converter 20 in which the number of repetitions of the processing component 22 is changed from the converter 20 having a structure in which the same processing component 22 is repeatedly connected. In one example, when the converter 20 has a structure in which the columns of the same processing component 22 are repeatedly connected, the generation unit 120 determines the processing component 22 according to the designation from the user who uses the converter 20. A converter 20 that converts the number of repetitions of the column is generated (S680, S690). In this way, when the user wants to increase or decrease the effect of the processing component 22 included in the specific converter 20, the generation unit 120 converts the number of repetitions of the column of the processing component 22 in accordance with an instruction from the user. can do.

  FIG. 7 shows an example of the processing flow of the genetic processing apparatus 100 in step S640 of FIG. The genetic processing device 100 executes the following processing in step S640 shown in FIG.

  First, the processing unit 140 executes the following processing in step S720 for each of the plurality of output data generated by each of the plurality of converters 20 stored in the converter storage unit 110 (S710, S730). . In step S720, the fitness level calculation unit 150 calculates the fitness level between the output data and the learning target data. When the fitness level calculation unit 150 finishes processing for all of the plurality of output data, the fitness level calculation unit 150 advances the processing to step S740 (S730).

  Subsequently, the selection unit 160 selects output data closer to the learning target data from among the plurality of output data based on the degree of matching of each of the plurality of output data (S740). The selection unit 160 may preferentially select output data having a higher matching degree from among the plurality of output data converted by the plurality of converters 20. The selection unit 160 may select output data whose fitness is higher than the standard fitness. In addition, the selection unit 160 may select output data in a range in which the fitness is predetermined from the top. In addition, the selection unit 160 may select output data at random under the condition that the output data having a higher degree of matching is selected with a higher probability.

  Subsequently, the selection unit 160 selects the converter 20 that generates the output data selected by the selection unit 160 as the converter 20 that converts the learning input data into data more suitable for the learning target data (S750). . When the process of step S750 is completed, the genetic processing apparatus 100 ends the flow.

  FIG. 8 is a diagram illustrating a converter 810 in which the same processing components according to the present embodiment are repeatedly connected and converters 820 and 830 in which the number of repetitions is changed. Within the converter 810, the same processing components indicated by black circles are repeatedly connected as 814A and 814B. When the integration unit 170 integrates the same processing components in the converter 810, the converter 810 is converted into the converter 820. Within the converter 820, the processing components 814A and 814B are integrated into the processing component 814C having a changed intensity. Further, the processing component 812, the processing component 816, and the processing component 818 in the converter 810 are not changed.

  Subsequently, when the dividing unit 190 divides the processing component 814 </ b> C in the converter 820, the converter 820 is converted into the converter 830. Within the converter 830, the processing component 814C is divided into processing components 814D, 814E, and 814F that have changed in strength. Here, the dividing unit 190 may divide all of the processing components integrated by the integrating unit 170 or may divide a part thereof.

  FIG. 9 is a diagram showing a converter 910 in which the same processing component sequence according to the present embodiment is repeatedly connected and a converter 920 in which the number of repetitions is changed. Within the converter 910, the same sequence of processing components is repeatedly connected as 912A and 912B. When the integration unit 170 integrates the processing component rows 912A and 912B in the converter 910, the converter 910 is converted into the converter 920. Within the converter 920, processing component columns 912A and 912B are integrated into the same processing component column 912C. Subsequently, when the dividing unit 190 divides the processing component column 912C in the converter 920 into the processing component columns 912A and 912B, the converter 920 is converted into the converter 910.

  FIG. 10 shows an example of a hardware configuration of a computer 1900 according to the embodiment of the present invention. A computer 1900 according to this embodiment is connected to a CPU peripheral unit having a CPU 2000, a RAM 2020, a graphic controller 2075, and a display device 2080 that are connected to each other by a host controller 2082, and to the host controller 2082 by an input / output controller 2084. An input / output unit having a communication interface 2030, a hard disk drive 2040, and a DVD drive 2060, and a legacy input / output unit having a ROM 2010, a flexible disk drive 2050, and an input / output chip 2070 connected to the input / output controller 2084. Prepare.

  The host controller 2082 connects the RAM 2020 to the CPU 2000 and the graphic controller 2075 that access the RAM 2020 at a high transfer rate. The CPU 2000 operates based on programs stored in the ROM 2010 and the RAM 2020 and controls each unit. The graphic controller 2075 acquires image data generated by the CPU 2000 or the like on a frame buffer provided in the RAM 2020 and displays it on the display device 2080. Instead of this, the graphic controller 2075 may include a frame buffer for storing image data generated by the CPU 2000 or the like.

  The input / output controller 2084 connects the host controller 2082 to the communication interface 2030, the hard disk drive 2040, and the DVD drive 2060, which are relatively high-speed input / output devices. The communication interface 2030 communicates with other devices via a network. The hard disk drive 2040 stores programs and data used by the CPU 2000 in the computer 1900. The DVD drive 2060 reads a program or data from the DVD 2095 and provides it to the hard disk drive 2040 via the RAM 2020.

  The input / output controller 2084 is connected to the ROM 2010, the flexible disk drive 2050, and the relatively low-speed input / output device of the input / output chip 2070. The ROM 2010 stores a boot program that the computer 1900 executes at startup and / or a program that depends on the hardware of the computer 1900. The flexible disk drive 2050 reads a program or data from the flexible disk 2090 and provides it to the hard disk drive 2040 via the RAM 2020. The input / output chip 2070 connects the flexible disk drive 2050 to the input / output controller 2084 and inputs / outputs various input / output devices via, for example, a parallel port, a serial port, a keyboard port, a mouse port, and the like. Connect to controller 2084.

  A program provided to the hard disk drive 2040 via the RAM 2020 is stored in a recording medium such as the flexible disk 2090, the DVD 2095, or an IC card and provided by the user. The program is read from the recording medium, installed in the hard disk drive 2040 in the computer 1900 via the RAM 2020, and executed by the CPU 2000.

  A program that is installed in the computer 1900 and causes the computer 1900 to function as the genetic processing device 100 according to the first embodiment includes a converter storage module, a generation module, a learning data storage module, a processing module, a fitness calculation module, and a selection module. , An integration module, a processing component storage module, and a division module. These programs or modules work on the CPU 2000 or the like to make the computer 1900 into the converter storage unit 110, the generation unit 120, the learning data storage unit 130, the processing unit 140, the fitness calculation unit 150, the selection unit 160, and the integration unit. 170, the processing component storage unit 180, and the dividing unit 190.

  The information processing described in these programs is read into the computer 1900, whereby the converter storage unit 110, the generation unit 120, the learning, which are specific means in which the software and the various hardware resources described above cooperate. The data storage unit 130, the processing unit 140, the matching degree calculation unit 150, the selection unit 160, the integration unit 170, the processing component storage unit 180, and the division unit 190 function. And the specific genetic processing apparatus 100 according to the intended use is constructed | assembled by implement | achieving the calculation or processing of the information according to the intended use of the computer 1900 in 1st Embodiment by these specific means.

  In an example, when communication is performed between the computer 1900 and an external device or the like, the CPU 2000 executes a communication program loaded on the RAM 2020, and based on the processing contents described in the communication program, the communication interface A communication process is instructed to 2030. Under the control of the CPU 2000, the communication interface 2030 reads transmission data stored in a transmission buffer area or the like provided on a storage device such as the RAM 2020, the hard disk drive 2040, the flexible disk 2090, or the DVD 2095, and transmits it to the network. Alternatively, the reception data received from the network is written into a reception buffer area or the like provided on the storage device. As described above, the communication interface 2030 may transfer transmission / reception data to / from the storage device by a DMA (direct memory access) method. Instead, the CPU 2000 transfers the storage device or the communication interface 2030 as a transfer source. The transmission / reception data may be transferred by reading the data from the data and writing the data to the communication interface 2030 or the storage device of the transfer destination.

  In addition, the CPU 2000 DMAs all or necessary portions from among files or databases stored in an external storage device such as the hard disk drive 2040, the DVD drive 2060 (DVD 2095), and the flexible disk drive 2050 (flexible disk 2090). The data is read into the RAM 2020 by transfer or the like, and various processes are performed on the data on the RAM 2020. Then, CPU 2000 writes the processed data back to the external storage device by DMA transfer or the like. In such processing, since the RAM 2020 can be regarded as temporarily holding the contents of the external storage device, in the present embodiment, the RAM 2020 and the external storage device are collectively referred to as a memory, a storage unit, or a storage device. Various types of information such as various programs, data, tables, and databases in the present embodiment are stored on such a storage device and are subjected to information processing. Note that the CPU 2000 can also store a part of the RAM 2020 in the cache memory and perform reading and writing on the cache memory. Even in such a form, the cache memory bears a part of the function of the RAM 2020. Therefore, in the present embodiment, the cache memory is also included in the RAM 2020, the memory, and / or the storage device unless otherwise indicated. To do.

  In addition, the CPU 2000 performs various operations, such as various operations, information processing, condition determination, information search / replacement, etc., described in the present embodiment, specified for the data read from the RAM 2020 by the instruction sequence of the program. Is written back to the RAM 2020. For example, when performing the condition determination, the CPU 2000 determines whether the various variables shown in the present embodiment satisfy the conditions such as large, small, above, below, equal, etc., compared to other variables or constants. When the condition is satisfied (or not satisfied), the program branches to a different instruction sequence or calls a subroutine.

  Further, the CPU 2000 can search for information stored in a file or database in the storage device. For example, in the case where a plurality of entries in which the attribute value of the second attribute is associated with the attribute value of the first attribute are stored in the storage device, the CPU 2000 displays the plurality of entries stored in the storage device. The entry that matches the condition in which the attribute value of the first attribute is specified is retrieved, and the attribute value of the second attribute that is stored in the entry is read, thereby associating with the first attribute that satisfies the predetermined condition The attribute value of the specified second attribute can be obtained.

  The program or module shown above may be stored in an external recording medium. As the recording medium, in addition to the flexible disk 2090 and the DVD 2095, an optical recording medium such as a CD, a magneto-optical recording medium such as an MO, a tape medium, a semiconductor memory such as an IC card, and the like can be used. Further, a storage device such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet may be used as a recording medium, and the program may be provided to the computer 1900 via the network.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the above-described embodiment. It is apparent from the scope of the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

  The order of execution of each process such as operations, procedures, steps, and stages in the apparatus, system, program, and method shown in the claims, the description, and the drawings is particularly “before” or “prior to”. It should be noted that the output can be realized in any order unless the output of the previous process is used in the subsequent process. Regarding the operation flow in the claims, the description, and the drawings, even if it is described using “first”, “next”, etc. for convenience, it means that it is essential to carry out in this order. It is not a thing.

20 converters, 22 processing components, 24 processing component groups, 100 genetic processing devices, 110 converter storage units, 120 generation units, 130 learning data storage units, 140 processing units, 150 fitness calculation units, 160 selection units, 170 integration unit, 180 processing component storage unit, 190 dividing unit, 810 converter, 812 processing component, 814 processing component, 814A processing component, 814B processing component, 816 processing component, 818 processing component, 820 converter, 910 converter, 912 Processing component array, 920 converter, 1900 computer, 2000 CPU, 2010 ROM, 2020 RAM, 2030 communication interface, 2040 hard disk drive, 2050 flexible disk drive, 2060 DVD drive, 2070 I / O chip, 207 Graphic controller, 2080 a display device, 2082 host controller 2084 output controller, 2090 a flexible disk, 2095 DVD

Claims (21)

A generation unit that generates a new converter by genetic processing from at least one converter that combines a plurality of processing components that process input data and output as output data;
An integration unit that integrates two or more processing components connected by connection between input and output in the new converter and replaces the processing components with processing components corresponding to the two or more processing components;
A dividing unit that divides one processing component of the plurality of processing components into two or more processing components;
With
The genetic processing apparatus, wherein the dividing unit divides the one processing component, which has been replaced by integrating the two or more processing components by the integration unit, into a set of two or more processing components. A generation unit that generates a new converter by genetic processing from at least one converter that combines a plurality of processing components that process input data and output as output data;
An integration unit that integrates two or more processing components connected by connection between input and output in the new converter and replaces the processing components with processing components corresponding to the two or more processing components;
With
Each of the plurality of processing components is a filter component that converts input data into output data,
The integrated processing unit integrates two or more of the same filter components connected by connection between input and output, and replaces the filter components with an integrated filter component that performs conversion with an intensity corresponding to the number of filter components to be integrated. . A generation unit that generates a new converter by genetic processing from at least one converter that combines a plurality of processing components that process input data and output as output data;
An integration unit that integrates two or more processing components connected by connection between input and output in the new converter and replaces the processing components with processing components corresponding to the two or more processing components;
With
Each of the plurality of processing components is a filter component that converts input data into output data,
The integration unit has the same two or more filter components when the conversion by the same two or more filter components connected by connection between input and output is the same as the conversion by the same one filter component. A genetic processing apparatus that replaces the same filter component with the same one. A filter storage unit that stores the identification information of the filter component in which conversion by the same two or more filter components coupled by connection between input and output is the same as conversion by the same one filter component;
The integration unit replaces the same two or more filter components with the same one filter component when two or more of the filter components stored in the filter storage unit are connected. Genetic processing equipment. A generation unit that generates a new converter by genetic processing from at least one converter that combines a plurality of processing components that process input data and output as output data;
An integration unit that integrates two or more processing components connected by connection between input and output in the new converter and replaces the processing components with processing components corresponding to the two or more processing components;
With
Each of the plurality of processing components is a filter component that converts input data into output data,
The genetic processing device, wherein the integration unit integrates repetitive portions where the columns of the filter components are repeatedly connected, and replaces the filter components with at least one filter component that performs conversion by the number of repetitions. The genetic processing device according to any one of claims 2 to 5, further comprising a dividing unit that divides one processing component of the plurality of processing components into two or more processing components.   The genetic processing device according to claim 6, wherein the dividing unit divides the one processing component that has been replaced by integrating the two or more processing components by the integration unit into a set of two or more processing components. The dividing unit divides the at least one converter into two or more partial converters by dividing one processing component of the at least one converter into two or more processing components,
The genetic unit according to any one of claims 1, 6, and 7, wherein the generation unit uses a partial converter of the two or more partial converters as at least a part of the new converter. Processing equipment.   The genetic processing device according to claim 1 or 8, wherein each of the plurality of processing components is a filter component that converts input data into output data. The genetic processing device according to claim 6, 7 or 9 , wherein the integration unit integrates two or more of the same filter components coupled by connection between input and output and replaces the integrated filter components with an integrated filter component.   The genetic processing device according to claim 10, wherein the dividing unit divides the integrated filter component into a set of at least two filter components that perform the same conversion as the integrated filter component together.   The integration unit integrates two or more filter components connected by connection between input and output, and uses a product matrix of conversion matrices of each of the two or more connected filter components as a conversion matrix The genetic processing device according to any one of claims 9 to 11, which is replaced with a component.   The generation unit further generates another converter in which the number of repetitions of the processing component sequence is changed from the converter having a structure in which the same processing component sequence is repeatedly connected. The genetic processing device according to any one of the above.   When the converter has a structure in which the same column of processing components is repeatedly connected, the generation unit repeats the processing component column in accordance with a designation from a user who uses the converter. The genetic processing device according to any one of claims 1 to 13, wherein the converter for converting the signal is generated. For each of the converters, a fitness calculation unit that calculates a fitness for conversion from learning input data to learning target data;
A selection unit for selecting the converter based on the number of processing components and the degree of fitness for each of the converters;
The genetic processing device according to any one of claims 1 to 14, further comprising:   The genetic processing device according to claim 15, wherein the selection unit selects the converter for each of the converters based on the fitness and the processing load of the processing component. A step of generating a new converter by genetic processing from at least one converter combining a plurality of processing components that process input data and output as output data,
Integrating two or more processing components connected by connection between input and output in the new converter by the integration unit and replacing the processing components with processing corresponding to the two or more processing components;
Dividing one processing component of the plurality of processing components into two or more processing components linked by connection between the input and output by the dividing unit;
With
A genetic processing method in which, in the step of dividing the one processing component, the one processing component replaced by integrating the two or more processing components by the integration unit is divided into a set of two or more processing components. A step of generating a new converter by genetic processing from at least one converter combining a plurality of processing components that process input data and output as output data,
Integrating two or more processing components connected by connection between input and output in the new converter by the integration unit and replacing the processing components with processing corresponding to the two or more processing components;
With
Each of the plurality of processing components is a filter component that converts input data into output data,
In the step of integrating the two or more processing components, an integrated filter that integrates the same two or more filter components connected by connection between input and output, and performs conversion with an intensity corresponding to the number of the filter components to be integrated. Genetic processing method to replace with parts. A step of generating a new converter by genetic processing from at least one converter combining a plurality of processing components that process input data and output as output data,
Integrating two or more processing components connected by connection between input and output in the new converter by the integration unit and replacing the processing components with processing corresponding to the two or more processing components;
With
Each of the plurality of processing components is a filter component that converts input data into output data,

In the step of integrating the two or more processing components, when conversion by the same two or more filter components coupled by connection between input and output is the same as conversion by the same one filter component, A genetic processing method for replacing two or more identical filter parts with the same one filter part. A step of generating a new converter by genetic processing from at least one converter combining a plurality of processing components that process input data and output as output data,
Integrating two or more processing components connected by connection between input and output in the new converter by the integration unit and replacing the processing components with processing corresponding to the two or more processing components;
With
Each of the plurality of processing components is a filter component that converts input data into output data,
In the step of integrating the two or more processing components, at least one filter component that integrates repeated portions in which the columns of filter components are repeatedly connected and performs conversion by the column of filter components with an intensity corresponding to the number of repetitions. Replace with the genetic processing method.   A program that causes a computer to function as the genetic processing device according to any one of claims 1 to 16.

Images