TW200831887A - Image processing apparatus, data processing apparatus, parameter adjusting method and record medium - Google Patents

Abstract

To provide a data processor, an image processor, a parameter adjustment method, and a program which can adjust parameters concerning data processing accurately with an easy manipulation. For making the adjustment of parameter values concerning image processing, a target data creating part 140 receives a correction directive to result data D2 from an operator and creates target data D3, then, a parameter adjustment making part 170 detects a parameter value which can make result data D2 agreed with the target data D3 within a predetermined tolerance and determines the value for an adjustment value.

Description

200831887 IX. Description of the invention: [Technical field to which the invention pertains] The technology of bedding processing (for example, parameters of image processing). The present invention relates to a parameter related to various adjustments (for example, a defect detection department [previous technique] in a printing workflow, inspecting a film, a plate, a printed matter of a 铷笠 不 分 分 土 2 2 (4) If there is a defect that is not expected, etc., it is called a check/printed matter f check"). The inspection/printed matter inspection is an image to be inspected by Putian (specifically, 'image data obtained by scanning a printed matter such as Ding Hao as a test object'), and as a result of its production, the original The image data of the benchmark is compared and the different aspects are detected as defects. In the inspection/printing inspection, the defects generated in the image to be inspected are detected by the processing of each recording, such as blurring, shaking, comparison processing, and isolation point removal processing. The job sensitivity of defects varies according to the environmental parameter values of each process. That is, in order to perform defect detection with appropriate sensitivity, it is necessary to appropriately adjust the parameter values in advance. The operation of this parameter adjustment was previously performed by each user based on experience and conducted by trial and error. Further, as a prior art of the inspection and inspection process, for example, a technique is considered in which the printed material to be examined is subjected to region division based on the image feature, and the attribute corresponding to the image region is used for each region. The parameter is subjected to the inspection process (refer to Patent Document 1). In this case, in order to perform appropriate defect detection for each image area, the parameter values must be appropriately adjusted. [Patent Document 1] Japanese Patent Laid-Open No. 2004-258470 No. 124049.doc 200831887 [Summary of the Invention] [Problems to be Solved by the Invention] However, in the inspection/printing inspection, a plurality of adjustments as described above are comprehensively performed. There are also many types of positive parameters. As a result, parameter adjustments require more time and labor. In addition, even when it is tried to adjust the parameters, when it is necessary to change the defect detection sensitivity rapidly, etc., there is no time to re-adjust and adjust the parameter values, and finally the inspection can be performed only by visual confirmation. The present invention has been completed in view of the above problems, and an object thereof is to provide a data processing device, an image processing device, and a parameter which can accurately adjust one or more parameters related to data processing (for example, image processing) with a simple operation. Adjustment methods and programs.乂 [Technical means for solving the problem] The invention of claim 1 includes: an image processing mechanism that performs specified image processing; and a target image producing unit that accepts a result image produced as a result of the image processing described above a correction instruction for producing a target image in which the correction instruction is reflected in the result image; and a parameter adjustment mechanism that adjusts a value of the parameter of the image processing to produce a result as a result of the image processing The image is identical to the above target image. The invention of claim 2 is the image processing apparatus of claim 1, wherein the image processing means performs image processing for comparing the reference image with the inspection target image obtained by outputting the current aesthetic image, and detecting Performing the above-mentioned defect portion generated in the image of the stomach image, and producing a result image in which the detected defective portion is displayed in a display form different from the non-defective portion, and 124049.doc -7- 200831887 describes image processing The parameters specify the detection sensitivity of the above defective portion. The invention of claim 3 is the image processing apparatus of the request item or the request item 2, wherein the target image producing unit displays the result image on the display surface, and receives the pair from the display surface. The input of the correction indication of the result image. The invention of claim 3 is the image processing device of claim 3, wherein the parameter adjustment mechanism _ surface causes the parameter value to change, and the surface image is determined whether the result image obtained under each parameter value is different from the target image. Therefore, the adjustment value of the above parameters is determined as the parameter value of the resultant image obtained by the target image. The invention of claim 5 is the image processing device of claim 4, wherein the plurality of gesturing means determines the adjustment value of the parameter as a set of parameter values obtained by obtaining a result image consistent with the image of the image The median. The invention of claim 4, wherein the image processing device of claim 4, wherein the parameterizing means determines the adjustment value of the parameter to obtain the target image in the over-rank of the parameter value 4 The value of the parameter when the result of the image is consistent. The invention of claim 7 is the image processing apparatus of claim 4, wherein the range specific mechanism is further included, and the range in which the parameter value is changed is specified based on the input set value. The invention of claim 8 is the image processing device of claim 4, further comprising: a variable, a specific mechanism, wherein the parameter value is changed according to the input setting value. The amount of change. The invention of claim 9 is the image processing apparatus of claim 3, wherein the 124049.doc 200831887 parameter adjustment mechanism uses a genetic algorithm to obtain a result consistent with the target image from a combination of more than one parameter value. The combination of the images determines the adjustment value of the above parameters as the values constituting the combination of the captures. The invention of claim 10, wherein the image processing device of claim 9, further comprising: an adjustment range specifying means for specifying a range in which the parameter value is changed based on the input set value. The invention of claim 11 is the image processing device of claim 9, further comprising: a change amount specifying means that specifies the amount of change when the parameter value is changed based on the input set value. The invention of claim 12 is the image processing apparatus of claim 2 or 2, further comprising: an adjustment target parameter specifying means, based on the specified feature amount obtained from the correction finger 2, from the parameter of the image processing The parameters that are the object of adjustment are specified. The invention of claim 13 includes a target image creation step of accepting, from the operator, a correction indication of the result image created as a result of the image processing, and causing the correction instruction to be reflected in the result image. a target image; and a parameter adjustment step of adjusting the parameter value of the image processing so that the resulting image produced as a result of the image processing is consistent with the target image. The invention of claim 14 is capable of realizing the following functions in the above computer by performing a computer image processing function for performing specified image processing; the target image creation function 'accepting the pair as the result of the image processing described above And a correction instruction of the produced result image, a target image for causing the correction instruction to be reflected in the result image; and a parameter adjustment function for adjusting a parameter value processed by the image 124049.doc 200831887 as the image The result of the processing is the same as the target image. The invention of the item 15 includes: a f-processing mechanism that performs a target data producing unit that accepts an instruction to correct the result obtained as a result of the above-mentioned information, and produces The above-mentioned repairs are not reflected in the target data of the above-mentioned results data; and the parameter adjustment mechanism: adjusts the parameter values of the above data processing to be processed as the above data: The data obtained is consistent with the above target data. The invention of claim 16 is the data processing device of claim 15, wherein the data processor 槿 && 'T ^ + wipes the image processing as follows: obtaining the reference image and outputting the reference image The inspection target image is compared with 'detecting the defective portion generated in the inspection target image, and image data for displaying the detected defective portion in a display form different from the non-defective portion is created and used as the result data. Obtained, the parameter of the above data processing defines the detection sensitivity of the above defect portion. The invention of the month-to-month item 17 is as in the request processing item 15 or the data processing device of the request item 16-/, the above-mentioned item; the beetle production organization displays the above-mentioned result data on the display: on the screen, 'accepted from the above display surface Input of correction instructions for the results data. The invention of claim 18 is the data processing device of claim 17, wherein the parameter adjustment mechanism-surface changes the parameter value, and determines whether the result data obtained under each parameter value is consistent with the target data, The value of (4) of the parameter is determined as the parameter value of the result data that is consistent with the above target data. The invention of claim 19, wherein the parameter adjustment means determines the adjustment value of the parameter as a set of parameter values obtained by obtaining result data consistent with the target data. Number of digits. The invention of claim 18 is the data processing device of claim 18, wherein the parameter adjustment unit determines the adjustment value of the parameter as a parameter when the result data in the process of changing the parameter value is initially obtained in accordance with the target data. value. The invention of claim 2 is the data processing device of claim 8, further comprising: an adjustment range specifying means for specifying a range in which the parameter value is changed based on the input set value. The invention of claim 22 is the data processing device of claim 18, further comprising: a change amount specifying means that specifies the amount of change when the parameter value is changed based on the input set value. The invention of claim 17 is the data processing device of claim 17, wherein the parameter adjustment mechanism uses a genetic algorithm to extract a combination of result data that is consistent with the target data from a combination of more than one parameter value. The adjustment value is determined to be the value of the combination that constitutes the capture. The invention of claim 24 is the data processing device of claim 23, further comprising: an adjustment range specifying means for specifying a range in which said parameter value is changed based on the input set value. The invention of claim 25 is the data processing device of claim 23, further comprising: a change amount specifying means that specifies the amount of change when the parameter value is changed based on the input set value. The invention of claim 26 is as set forth in claim 15 or claim 16 of the data processing apparatus 124049.doc -11 - 200831887, which includes an adjustment object parameter specific mechanism based on the designation obtained from the above correction instruction The feature quantity specifies a parameter that becomes an adjustment target from the parameters of the above data processing. The invention of claim 27 includes: a target data creation step, wherein (4) the author accepts an instruction to correct the result obtained as a result of the data processing, and produces a target data for reflecting the correction instruction in the result: And the parameter adjustment step 'which adjusts the parameter values of the above data processing so that the data obtained as a result of the above data processing is consistent with the above target data. The invention of claim 28 is implemented by the computer by means of a computer to implement the function of the above-mentioned computer: the f-processing function 'which performs the specified data processing; the target material production function, which accepts the contradictory, +, and Further, as a result of the correction of the result data obtained as a result of the above-mentioned poor material processing, the production of the above-mentioned correction instruction is reflected in the target data of the above-mentioned, and the fruit and the greed; and the parameter adjustment function is adjusted. The parameter values of the second rationale are such that the information obtained as the result of the above-mentioned poem processing (4) is consistent with the above target data. [Effects of the Invention] According to the invention of the request item 13, 14, the operator does not directly input the parameters to hunt! ί Parameter adjustment for the correction indication of the resulting image. I7, you can adjust the parameters related to the drawing and processing with simple operation. Also, because of swallowing. As a result of image processing, the image of the image that is served is consistent with the target image, so that the parameter value of the image processing result brought by the operator is obtained. That is, the parameters of the image processing can be accurately adjusted. According to the inventions of claims 2 and 16, it is possible to adjust the parameters of the detection sensitivity of the defective portion generated in the image of the inspection target 124049.doc • 12. 200831887 early and accurately. According to the invention of claim 3, the input of the correction instruction to the result image is accepted from the display screen, so that the operator can input the correction instruction for the result image simply and accurately. As a result, the parameters can be adjusted more easily and accurately. According to the invention of claim 4, the parameter value is changed - the parameter value of the face detection is appropriate (that is, the parameter image obtained by the parameter value is consistent with the target image ^), so that the appropriate detection can be accurately performed. The parameter value. According to the invention of claim 5, the adjustment value of the parameter is determined as the median of the set of parameter values of the resulting image that is consistent with the target image, and thus the optimum value can be determined from the appropriate parameter values. Adjust the value for the parameter. W. According to the invention of claim 6, the parameter adjustment value is determined as the parameter value when the result image matching the target image is first obtained in the process of making the parameter value in the process of activating the parameter value, so that the parameter adjustment can be quickly performed. C- According to the invention of claims 9 and 23, the genetic algorithm is used to extract the combination of the best parameter values, so that the parameter adjustment can be performed quickly. According to the invention of claims 7, 10, 21, and 24, the range in which the parameter value is changed is specified based on the input set value, so that the parameter adjustment range can be narrowed. Thereby, the parameter adjustment can be performed quickly. According to the invention of claim 8, U, 22, and 25, the amount of change when the parameter value is changed is specified based on the input set value, so that the parameter value obtainable in the parameter adjustment range can be reduced. Thereby, the parameter adjustment can be performed quickly. According to the invention of claim 12, according to the specification of the self-correction instruction, 124049.doc -13-200831887, the special government of the government, adjusts the object parameters, and therefore A has gone from the point where you are in order to obtain the target map. Like one, the parameters can be adjusted quickly and accurately. ...numbers. According to the invention of claims 15, 27, and 28, the operator can perform parameter adjustment for the correction instruction of the result data. "’ The number of tables related to data processing and adjustment. In addition, the data obtained from the results of :::: material processing is in accordance with the target: the reference value of the data processing result expected by the operator is obtained. That is, the parameters of the data processing can be accurately adjusted. According to the invention of claim 17, the self-displaying dry gold .τ . A β accepts the result data from the self-image screen: the operator can simply and accurately input the result indication. As a result, the parameters can be adjusted more easily and accurately. In the invention of 8, the surface value is changed, and the surface value is detected (ie, the result data obtained under the parameter value and the target lean material-induced parameter value), so that the appropriate parameter value can be accurately detected. . According to the invention of claim 19, the adjustment value of the parameter is determined as the median of the set of parameter values obtained by obtaining the result data with the target data, and therefore the optimum value may be determined as the optimum value in the appropriate parameter value. The adjustment value of the parameter. According to the invention of claim 20, the parameter adjustment value is determined as the parameter value when the result data matching the target data is initially obtained in the process of changing the parameter value, so that the parameter adjustment can be quickly performed. According to the invention of claim 26, the specified feature quantity 'specified adjustment target parameter obtained from the self-correction instruction is selected, so that the type of the parameter to be adjusted can be appropriately selected in order to obtain the result data related to the target data 124049.doc 200831887 . This allows the parameters to be adjusted quickly and accurately. [Embodiment] [First Embodiment] Fig. 1 is a view showing a configuration of a printing system 100 including a defect inspection device 1 as an aspect of a data processing device according to a first embodiment of the present invention. The defect inspection device 1 is a device that detects defects generated in a printed matter or the like produced as described below, and has a function of adjusting image processing parameters related to the defect detection process. <1-1. Printing system configuration> The printing system 100 is a defect inspection device 1, a printed material production shelf 2, a plate-making device 3, an output device 4, and an image scanner 5 via a LAN (L〇cal Area Network) A system in which the network 1 is electrically connected to each other. In the printing system 100, a series of workflows from the production of printed materials to the printing of plates and outputs are carried out. The defect inspection device 1 is a device for detecting a check/printed matter of a defect generated in a printed matter, a plate-making film, a printing plate, or the like obtained by outputting a reference image. In the following, the printed matter, the plate-making film, the printing plate, and the like which are the inspection objects for the inspection/printed matter inspection are collectively referred to as the "r inspection object". The inspection/printing inspection refers to the image data of the inspection object (hereinafter referred to as "inspection target data D1") and the image data which is the basis for the inspection object (hereinafter referred to as "reference data D". The difference is detected as a defect. 124049.doc •15· 200831887 ▲Printed data production device 2 is a device for producing printed materials. More specifically, 'the image of the printed image is placed in a layout process to create layout data, and the RIP (Raster Image Process) processing (raster processing) is used to convert the created layout data. For multi-gray image data. Further, it is also possible to adopt a configuration in which RIP processing is performed on an independent device different from the printed material producing device 2. The print body making device 2 can generate image data corresponding to various resolutions for use by the same layout material. For example, a dot image image having a resolution of about 2400 dpi is generated as image data for output, and on the other hand, a plurality of grayscale image data having a low resolution of about 300 dpi can be generated for inspection. Benchmark data D〇. Preferably, the resolution of the reference data D〇 is determined based on the inspection target data ,1, and in this embodiment, the resolution of the reference data D0 is set to the same level as the inspection target data m. For example, when the inspection target document 01 is obtained by the digital camera photographic print having a resolution of 300 dpi, the printed material creation device 2 C performs RIP development on the layout data with a resolution of 3 〇〇 dpi to create the reference material D0. The prepared reference data D0 is transmitted via the network N to the defect inspection device 1 for inspection/printing inspection. The plate making apparatus 3 is a device for outputting a printing plate based on the dot image data (MCTP (ComPuter t0 Mnt)). More specifically, a printed image based on dot-spot image data is formed on the plate by laser exposure, thereby producing a printing plate. Furthermore, it is also possible to use the image typesetting machine to make a plate-making film based on the dot image data, and to make a plate using the prepared plate film. In this case, the spoon 124049.doc -16· 200831887 in the plate making apparatus 3 includes an image typesetting machine. The output device 4 is a device that prints printing paper using a printing plate produced by the plate making device 3. Further, it is also possible to digitally output the printing paper directly from the dot-coded image without passing through the printing plate. ~ The scanner 5 is a device that reads the inspection object/printed matter inspection object and obtains the inspection target data (1). The acquired inspection target data (1) is transmitted to the defect inspection device 经由 via the network N, and is used for pattern inspection/printed matter inspection 〇c <1-2·construction of the defect inspection device> Realized by the computer. More specifically, the defect inspection apparatus 1 mainly includes a control unit U including a CPU (Central Processing Unit) Ua and R〇M (Read 〇nly).

Memory ‘only memory】llb, and RAM (Rand〇in Access

Memory, random access memory) Uc, which realizes the following functions; the memory unit 12' stores a program P for causing the computer to function as the defect inspection device 1, and the like; the operation unit 13 includes operations for operating a mouse or keyboard for inputting various instructions, a display portion 14 such as a display, and a portion 15 including a hard disk or the like for reading and writing data between the media reader 151 and the recording medium; As an interface communication unit 丨6, it is used for data transfer between other devices on the network ν'. Further, in the defect inspection device 1, a so-called GUI (Graphical User Interface) that can be processed while the operation content of the operation unit 、3 or the processing status of various processes is displayed on the display unit i4 can be performed. 101 (see FIG. 2) is realized by the functions of the control unit 11, the operation unit 13, and the display unit 124049.doc -17·200831887 14. Thereby, the operator can use the operation unit 13 ♦ ‘, 员不# to provide an instruction while referring to the displayed screen. The trick is to make a computer <1-3·related to the adjustment of the parameters> defect inspection device! It has the function of adjusting the image processing number of the check/print check. Fig. 2 is a diagram for explaining the configuration related to the parameter adjustment function. The configuration device 1 includes a reference beetle acquisition unit 110, an inspection target data acquisition unit 12 (), a defect detection processing unit 130, a target data creation unit 140, and an adjustment target parameter identification unit 15A. The adjustment constant determination unit 16A and the parameter adjustment processing unit 17A. The components are implemented by the control unit 11 by the specific program p stored in the storage unit 12. The reference data acquisition unit 1 acquires the reference material D0 based on the privacy of the operator received via the operation unit 13. More specifically, the reference material D is acquired from the printing material creation device 2 via the network N. Further, as shown in FIG. ,, it can also be read and recorded on the portable recording medium M by a media reader 151 including an M〇 driver or a CD-R/RW drive (for example, MO (magnetic optical disC'). The reference data D〇 is obtained from the reference data D0 of CD-R (recordable compact disk)/RW (Rewritable compact disk). The reference data acquisition unit 110 stores the acquired reference data D〇 in the storage unit 12. The inspection target data acquisition unit 12 receives the inspection target data D1 based on the instruction received from the operator via the operation unit 13. More specifically, the inspection target data D1 is acquired from the image scanner 5 via the network N. In addition, the recording medium M of the inspection target data D1 obtained by the photographing object to be photographed by a digital camera such as a digital camera can be read by the media reader/writer 151 to obtain the inspection target data D1. The inspection target data acquisition unit 120 stores the acquired inspection target data D1 in the storage unit 12. The reference data D0 is illustrated in Fig. 3(a). Further, in Fig. 3(b), the inspection target material D1 of the inspection object produced based on the reference data DO illustrated in Fig. 3(a) is exemplified. As described above, the reference data DO is the reference for the production of the printed matter (the image data, and the inspection target data D1 is the image data of the inspection object. Preferably, the inspection target data D1 and the reference data D0 are completely identical. Due to various factors (such as dust or misalignment) which may occur in the printing step, as shown in FIG. 3, it is possible to generate a portion (defect) different from the reference material D0 in the inspection target data D1. The defect detection processing unit 130 detects the defect by executing the following defect detection algorithm. The defect detection processing unit 130 performs image processing for storing the reference data D〇 stored in the memory unit 12 and outputting the reference data DO. The obtained inspection target data D1 is compared, and the defective portion generated in the inspection target data D1 is detected, and the result data D2 indicating the detected defective portion is displayed in a display form different from the non-defective portion. More specifically, , the defect data stored in the reference data DO and the inspection target data D stored in the A memory department 12 is tested to detect the inspection object Defects generated in 〇1. The so-called defect detection image processing, more specifically refers to blur processing, sloshing processing, comparison processing, and isolated point removal processing, which are used for detecting defects-series calculus The method (hereinafter referred to as "defect detection algorithm") 124049.doc -19·200831887 is performed in a specific order. The defect detection processing unit 13A is provided with a paste processing unit 13 as a process for performing defect detection image processing. The 丨, sway processing unit 、, the comparison processing unit 33, and the isolated point removal processing unit 134. The result is a result image creation unit 135 that reflects the detection result creation result data D2. The blur processing 邛13 1 is based on the parameters of the blur processing (blurring) The blur processing is performed on the parameter p 1). The blur processing is performed on the reference data D0 and the inspection target data D1; the processing of the blur, more specifically, the blur processing is performed as follows: the pixel of the processing target and the pixel located therein are calculated The average value of the pixels in a specific number nearby, and the calculated value is set as the output value of the pixel. Here, the degree of blur is calculated by the average value. The pixel range (more specifically, σ is the blur radius) is determined, and the blur radius is set as the fuzzy parameter Ρ 1. By combining the blur processing into the defect detection algorithm and performing the blurring process, the blur conditions of the same degree can be appropriately compared. The sway processing unit 132 performs a swaying process based on the parameter of the swaying process (the swaying parameter Ρ2). The swaying process refers to the assumption of the arrangement position of the image data of any one of the reference data D〇' inspection target data D1. The process of shifting and comparing the two images is performed, and the range of the swaying pixels (more specifically, the sway radius of the non-moving pixel range) is set as the comparison parameter P2. By swaying the sway processing into the defect detection algorithm and The sloshing process is performed, and even when the arrangement position of the line image or the pattern of the inspection target data D1 is shifted from the arrangement position of the reference data ( (the case where the pixel shift occurs), the pixel shift can be eliminated and detected. The original difference value. The comparison processing unit 133 performs 124049.doc -20-200831887 than the rutting process based on the parameter of the comparison processing (comparison parameter p3). The § stomach comparison processing is a process of detecting the gray level difference generated by the reference data training and the inspection target data 1 as a defect. More specific yet. For the reference body material D0 and the inspection target data D1, the difference value of the gray level difference is calculated in units of each corresponding pixel, and when the calculated difference = greater than a certain threshold value, the pixel is determined as defect. The threshold (i.e., grayscale edge) that determines the grayscale difference that should be detected as a defect is set to the comparison parameter P3. The isolated point removal processing unit 134 performs the isolated point removal processing based on the parameter of the isolated point removal processing (the isolated point removal parameter P4). The so-called outlier removal processing is performed by removing the isolated point from the reference data D〇 and the inspection target data D丨 when the pixel is isolated in a small pixel range. Here, the pixel range determined to be an isolated point (more specifically, the number of isolated point removal pixels defining the number of pixels forming the isolated point) is set as the isolated point removal parameter P4. Further, in the defect detection algorithm of the embodiment, the processing units 131 to 134 perform specific image processing on the reference data and the inspection target data D1 read by the storage unit 12, and more specifically, For example, the following procedure is performed. First, the blur processing unit 131 performs blurring processing on the reference material do and the inspection target data 〇1. By performing the blurring process, the reference data D0 of the image data obtained by the RIp expansion and the inspection target data D1 obtained by the camera or the like are corrected to the same degree of blur. Then, the sway processing unit 132 performs sway processing on the inspection target data 〇1, and the comparison processing unit 133 performs comparison processing on the inspection target data 124049.doc-21/200831887 that has been swayed, and the reference data DO, thereby eliminating the pixel deviation. After the shift, a meaningful grayscale difference is detected. Further, the isolated point removal processing unit 134 performs an isolated point removal process to remove redundant isolated points generated in the result of the comparison processing. In this way, it is avoided that small isolated points that are not to be detected as defects are also detected as defects. Referring again to Figure 2. The result image creation unit 135 creates the result data D:2. The result data D2 characteristically displays the defective portion in the inspection target data D1 in a display form different from the non-defective portion (that is, the portion other than the portion captured as the defect) (that is, the reference data 〇〇 and the inspection object) The data (1) is the result of the defect detection algorithm, and the image data of the object data is referred to as the defect. The result data D2 obtained by performing the defect detection algorithm on the reference data and the inspection object data D1 illustrated in Fig. 3 is illustrated in Fig. 4(a). In the result data D2, the non-defective portion is displayed lightly, that is, the region (appropriate region SI, S2, S3) which is determined not to have a meaningful difference from the reference data D 中 in the inspection target data D1, and the defective portion, That is, the region (defective regions T1, T2) which is determined to have a meaningful difference from the reference feedstock D〇 is displayed in a specific color (for example, magenta) after being fulled by the fullness processing. The target data creating unit 140 accepts the operator's correction instruction for the result data ’ 2 and creates image data (hereinafter referred to as "target data D3") which is reduced to the index I data D2. β) 中彳] indicates the target data D3 produced by the operator instructing the correction of the result data D2 illustrated in Fig. 4(a). For example, in the case of the operator (4) who received the defect area in the defect area T1, T2 of Fig. 4(4) 124049.doc -22- 200831887 2: fruit data D2, the target data production department The production shown in the standard data ^3 corrects the defect area Τ2 to the appropriate area. More specifically, it is corrected that the defect area of the result data D2 can be widely displayed as a state in which the appropriate area is displayed lightly. In addition, the operation of the repair and display *Yes, 纟#果资_ on the screen provides the result data 〇2 调2不不€€ Figure 7). This is further specified below.

The parameter of the C PI target parameter specifying unit 150 from the defect detection image processing 敫 the centering characteristic becomes one or more parameters of the adjustment target (hereinafter referred to as "tuning parameter"). The specificity of the object parameters is adjusted, and more specifically, the hunting is performed by accepting parameters that are selected by the operator to be input in the parameters IM to P4. The positive derivation i7 number determining unit 160 determines the "adjustment width f" j "the number of divisions" of each adjustment target parameter. The decision to adjust the width and the number of divisions is more specific and the Z straw is carried out by the adjustment width and the division value of the operator input by the operator. The term "adjustment width" refers to the adjustment range of the adjustment target parameter (that is, the parameter value range I in which the parameter adjustment processing unit 170 described below changes) is defined by the minimum value and the maximum value of the parameter value. The "knife number" refers to the change width of the parameter value (that is, the amount of change when the parameter adjustment processing unit 170 changes the parameter value). "For example, the adjustment width of the blur parameter ^ is set to "maximum value = 8" Pix "SJ value 1 pix", when the number of divisions is set to "1 pix" (see Fig. 9), the parameter adjustment processing unit 使7〇 sets the value of the blur parameter p丨 from "^ P1X" to "8" Pix" changes in the unit of "丨Pix", and the value of the fuzzy parameter P1 is best determined by 124049.doc -23- 200831887 2. In addition, the width of the adjustment is adjusted by the width of the division number. The parameter value that can be obtained when the adjustment range of the control is changed is called the "trial value". That is, the adjustment value of the parameter is determined by any one of the trial values. Furthermore, when "〇" is set to the minimum value When the parameters are adjusted:: It is also possible to decide Is "°". This case, in a substantially continuous defect detection method is not exclusive for the parameter value is set to adjust a "G" of the process.

The C/parameter adjustment processing unit 170 determines an adjustment value of the adjustment target parameter. parameter

The adjustment processing unit 17G includes a parameter value-adjusting factor determination unit and a parameter value determining unit 172. The N parameter value appropriateness determining unit 171 sequentially changes the parameter value (more specifically, the change width determined by the number of divisions is changed by the adjustment width to determine the adjustment range of the parameter value (in other words, The parameter value group 0, P2, P3, P4) of the parameter P丨~P 4 of the defect detection image processing (ie, 'parameter^, combination of values) causes the parameter values of the adjustment object to change sequentially. )), whether the result data D2 obtained by judging each parameter value is consistent with the target data D3. That is, it is judged whether or not each combination is a combination of parameter values of the result data D2 which is provided in association with the target material D3 for all combinations of the parameter 卩丨~ each of the values. Among them, the judgment as to whether or not the target data D3 coincides with the result data D2 is more specifically determined by judging whether or not the defective area portions of the two image data are identical. In addition, the parameter value of the result data D2 determined by the parameter value appropriateness determination unit 171 to be identical to the target data is referred to as "appropriate value". In addition, the parameter value group which provides the result data D2 which is the result of the target data D3 is called 124049.doc -24- 200831887 "appropriate parameter value group". In other words, when the parameter parameter "1" is adjusted in the parameter_4, the value of the adjustment target parameter P3 is set to "1". The defect detection algorithm is executed. Thus, the result obtained in the (4) = ^ trap region portion and the defect region portion of the target data (1) - ▲ : : : The value group is called the appropriate parameter value group. X, for parameter P3 and 5, the parameter value n" can be referred to as an appropriate value. : The value ΆΚΠ 72 determines the value (that is, the adjustment value) that is output as the adjustment result of the adjustment target parameter. More specifically, the adjustment value is determined as the set median of the appropriate parameter value group. That is, when the median of the set consisting of the appropriate values is appropriate and is determined as the adjustment value, if the object parameter is one type, as shown in Fig. 11 (4), == a constitutes the median of the set Q1. The trial value b is determined and determined as the adjustment value. <2·Processing> Next, the defect inspection device! The parameter adjustment processing will be described. <Flow of the entire parameter adjustment processing> a Fig. 5 is a flowchart showing the overall adjustment of the parameter adjustment processing of the defect inspection apparatus 1 (more specifically, the adjustment processing of the defect detection image processing). First, the reference data acquisition unit 110 and the inspection target data acquisition unit 120 respectively acquire the reference data DG (see (4) 3 (4)) and the inspection target f (see FIG. 3(b)) and store the information in the storage unit 12 (step s). Then, the defect detection processing unit 13 refers to the reference resource obtained in step S1 124049.doc •25- 200831887

Material DO and inspection object data D oblique parameter-under-defect defect detection algorithm, detecting and detecting defects of f-like material m, and making reflection detection D2 (refer to FIG. 4(a)) (step S2) e ', , ° He Beiji, the target data production department 14. The operator receives the result data and shows the target content, and produces the target content D3 (see Fig. 4(b)) (step S3). Then, the adjustment target parameter specifying unit 15 selects the specific adjustment target parameter from the defect detection image processing parameters P1 to P4, and determines the adjustment width and the number of divisions of each _ (step S4). I. The parameter adjustment processing unit 170 determines the adjustment target (step S5). The adjustment value is then the parameter adjustment processing unit 170, which determines the upper value determined in step 85, the target data created in step S3, and the result data obtained when the defect detection algorithm is executed for the determined 6 weeks. The parameter value 14 is displayed (step S6). When the display unit parameter adjustment processing unit m receives YES in the instruction step S7 (hereinafter), the adjustment target is small (step 乂 double value 〇 is determined as the adjustment value determined by the step (step S8). The above indication indicates that the operator of the result of the shovel S 6 is the result of setting the adjustment target parameter value to the adjusted value of 4, thereby ending the adjustment processing of the parameter value. Otherwise, step S3 S4 Further, the processing of step S3 will be described more specifically. The processing of step S3 will be described more specifically. Fig. 6 is a flowchart of the processing of the table. Doc -26-200831887 First, the target asset creation unit 140 displays the acceptance screen G of the correction instruction of the result data D2 on the display unit 14 (step su). The figure shows a configuration example of the acceptance screen G. The acceptance screen G includes: a region y which displays the entire result data D2 to be the correction target; a region g2 which enlarges and displays the region to be corrected in the result data D2; and an area which displays a selection panel for determining the correction content. Data system The portion 140 accepts the designation of the region to be corrected in the region (step S12). That is, the operator observes the entire result asset 2 displayed in the region gl, and judges whether or not there is a portion to be corrected in the result data D2. In the case of the correction, the region to be corrected is designated by the drag operation of the mouse. The target data creation unit 140 accepts the region finger $. Further, it is also possible to use various finger devices other than m (for example, ' Execution of the ball, touch pad, input board, etc.) for area designation.

C - After receiving the area to be corrected in the area gl, the target data creating unit 140 will continue to use the location A. Then, the area of the "Haohai" area is enlarged and displayed in the area g2 (step S13). For example, if the operator does not specify the area u〇 in the result data D2 displayed in the area: as shown in FIG. 7, the part 140 accepts the regional training and the private area will be within the E field U0. The enlargement is displayed in the area g2 ° = 'the target data creation unit 14 〇 the acceptance area, and the designation correction instruction g (7) is referred to as "correction target area U") (step S14). P, by the operator dragging the result data in the g2 (4) as 4 to more clearly specify the area, "the part to be corrected, the target data production department 140 is connected to 124049.doc -27- 200831887 The specified area is used as the correction target area. The result is shown as: _ a part of the area, so the area in operation refers to ^: and the clearer area designation is made. After receiving the designation of the correction area _, the target accepts the amendments to the correction target area U. _ Γ

S15). More specifically, the input operation is performed by the operator selecting and clicking the panel "Add" or "Delete J" to input the correction content in the selection panel displayed in the area (2). - After receiving the input of the correction content The target data creation unit "0 corrects the content indicated in the data D2 (step S16). For example, the operator selects the "deletion" panel, and the target data creation unit MO includes the correction target area U. The defect area is corrected to the appropriate area. When the operator selects the "Add" panel, the target data creation unit (10) corrects the appropriate area included in the correction target area U to the defective area. Hereinafter, an area in the correction target area U that is actually changed to a defect area or an appropriate area is referred to as a "differential area V". Further, in the above, upon receiving the designation of the correction target area !^ in the region g2 (step S14), the difference region 乂 included therein is corrected (step S16), and the difference region can be directly accepted in the region g2. Designation. At this h Φ, the operator can provide, for example, an indication of correcting one of the appropriate areas or defective areas displayed in the area g2 to the defective area or the appropriate area (e.g., the correction instruction E3 of the reference picture 3). Then, the target data creating unit 14 displays the result of the correction in step (5) reflected in the result data D2 in the area g (step s7). 124049.doc -28- 200831887 ie the display of the change difference area v is shown in the area §1

In the case of the C object area, the target f material creation unit has just performed the processing of steps S12 to S17 again and accepts the correction instruction for the result data D2. On the other hand, when the operator judges that the instruction to turn on all the correction parts has been completed, the "Details" panel is selected and an instruction to end the correction instruction is input. Then, the target data creating unit i40 waits for the instruction input of the correction completion (the step of the instruction input operation is more specifically performed by the following: • The operator selects and clicks on the selection panel displayed in the area g3 to input the following instruction. Face i "Details" 'The main purpose of this instruction is to end the correction instruction, and then move to the page for setting the details of the parameter to be adjusted. The nuclear author observes whether the image of the area g In the case of the step-correction part, when the part to be corrected is observed, the "details" panel is not selected, but the input of the correction area to the end of the correction instruction is again designated in the area _ (in step S18) After γΕ8), the target poor material preparation unit 140 completes the creation of the target material D3, and stores the created target data D3 in the storage unit 12 (step S19), thereby ending the processing of the step “2-2·Adjustment Specific Processing of Object Parameters> The processing of step S4 will be more specifically described. Fig. 8 is a flowchart showing the processing of step μ. If the processing of step S3 is completed (i.e., the acceptance is performed) After receiving the operation of the "Details" panel, the adjustment target parameter specifying unit 15 displays the acceptance information of the setting input of the parameter adjustment on the display unit 14 (step S2). Fig. 124049.doc -29-200831887 9 A diagram showing a configuration example of accepting a facet. The acceptance screen η includes a region hi for accepting selection input parameters, a region h2 for accepting an input of the adjustment width and the number of divisions, and a selection panel for determining and eliminating the display setting input. Area h3.

The adjustment target parameter specifying unit 150 first accepts the selection of the parameter to be adjusted in the region hl (step S22). More specifically, the type of the parameter of the defect detection image processing is displayed in the area ", and the check box of each parameter P1 to P4 is displayed. The operator can select the desired adjustment parameters from the parameters p 1 to P4 by operating the mouse to check the check box. In the example of FIG. 9, the "blur radius" item indicating the blur parameter..., the "shake radius" item indicating the sloshing parameter P2i, and the "number of isolated pixels removed" parameter of the isolated point removal parameter P4 are displayed in the area hl. . Further, although it is in a hidden state in Fig. 9, if the lever displayed on the right side of the area " is dragged, an item indicating the "gray edge" of the comparison parameter ?3 is displayed. The operator can select the parameters of the desired adjustment by checking the check box of the desired parameters. When the operator checks the arbitrarily-check box and selects a parameter, the object parameter is adjusted; the t unit 150 accepts the input of the adjustment width and the number of divisions selected by the (4) S22 in the region h2 (step S23). More specifically, the input box of the adjustment width (that is, the most 'large value disc: value of the adjustment range>) and the input box of the division number are displayed in the area h2. The operator can input the desired adjustment width and the number of divisions by rounding the values for each rain frame. As shown in Fig. 9, when the operator nucleates the "blur radius °" item and selects the fuzzy parameter P1 as the adjustment target parameter, the adjustment object 124049.doc • 30 - 200831887 parameter specific part 150 pairs "blur radius" The display portion is shaded, and an input box for adjusting the width of the blur parameter ^ and an input box for the number of divisions are displayed in the area h2. In Fig. 9, the operator inputs the maximum value "8 (Pix)" and the minimum value "1 (pix)" as the adjustment width of the blur parameter ρι and inputs "l(pix)" as the division number. Then, the adjustment target parameter specifying unit 15 waits for input of the selection of the completed adjustment target parameter and the instruction of the adjustment width of each adjustment target parameter and the setting input of the division number (step S24). More specifically, the instruction input operation is performed by the operator selecting and clicking the panel r of the instruction to input the detailed setting of the parameter setting in the selection panel displayed on the area h3. When the operator judges that there is still a parameter to be supplied to the adjustment target, the parameter is further selected in the area !^ without selecting the "Decision" panel. In this case, the adjustment target parameter specifying unit 150 performs the processing of steps S22 to S24 again and accepts the detailed setting of the input. On the other hand, when the operator decides to complete the detailed setting of the parameters, select the "Decision" panel and enter the instruction to end the detailed setting. After the target parameter specific unit 丨5〇 receives the instruction input of the detailed setting end (YES in step S24), the parameter checked in the check box is read as the parameter of the rounding object (step S25), and then read. The adjustment width and the number of divisions of the adjustment target parameters are entered (step S26). Furthermore, if the operator clicks the cancel panel displayed in the operation area h3, the parameter can be reselected or the adjustment width and the number of divisions can be input. When the operation cancels the h-shape of the panel, the adjustment object parameter specifying portion 150 invalidates the previous operation of the operator. This completes the processing of step S4. 124049.doc • 31· 200831887 <2_3. Decision processing of adjustment value> The processing of step S5 will be more specifically described. Figure ΐ() shows the processing flow of step ^. After the processing of step S4 is completed (that is, after the acceptance screen accepts the operation of the "decision" panel), the acceptance panel is displayed again on the display portion; After receiving the operation of the "Recalculation" panel in the face G, the process of the step Μ is started. The initial values of the numbers P1 to P4 are obtained (step S31). More specifically, the parameter adjustment processing unit 170 sets the parameter of the defect detection image processing. The parameter value of the adjustment target parameter in the parameters P1 to P4 is set as the minimum value of the adjustment width set for the parameter of the surrounding object. Furthermore, the value of the current setting is set to the initial value for the parameter that is not the parameter to be adjusted. The value of the parameter that is not the adjustment object is always fixed at the initial value. Hereinafter, the parameter value group composed of the initial values of the parameters ρι to ρ42 will be referred to as "initial parameter value group". Then, the defect detection processing unit 13 sets the set parameter value (that is, the initial parameter value group) as the determination target parameter value group, and the reference data acquired in step S1 in the determination target parameter value group. The inspection target data is executed to execute the defect detection algorithm, and the result data D2 reflecting the detection result is created (step S32). Then, the result data D2 created in step S32 is stored in the storage unit 12 (step S33). Then, the parameter value appropriateness determining unit 171 compares the result data D2 created in step S32 with the target data D3 created in step S3, and determines whether the defective area portion of the result data D2 is related to the target data 124049.doc -32· 200831887 The defect area VIII A 4 points are consistent (step S34). More specifically, in the result data D2, the part of the jAt A.», 苟 defect area is the defect area in the target data D3 and the target; I;»: owe, the knowledge of the material D3 does not exist. In the case of the defect area (ie, the gentleman ^^ 丨立丨, the Ό 料 material D2 and the defect area contained in the target data D3 are exactly the same h-shaped), the judgment result data D2 is consistent with the target data D3. In other cases, it is judged that the two image data are inconsistent. In step S34, when it is determined that the defective area portions of the two image data are identical, the parameter value appropriateness determining unit 171 sets the parameter value used in the defect ^ algorithm performed in step S32 (in the adjustment target parameter) In the case of a plurality of U-shapes, a combination of parameter values used in the defect detection algorithms is determined to be an appropriate value, and the parameter value (when the adjustment target parameter is a plurality of If-forms) is such The combination of the parameter values is stored in the memory unit ^ (step S35). In other words, the determination target parameter value group used in the defect detection algorithm executed in step S32 is determined as an appropriate parameter value group (that is, each value of the adjustment target parameter constituting the determination target parameter value group is determined to be an appropriate value), Each value of the adjustment target parameter determined to be an appropriate value is stored in the memory unit 12. In the other aspect, in the step S34, when it is determined that the defective area of the two image data or the four-knife shape is not formed, the parameter value appropriateness determining unit 丨7丨 adopts the defect detecting algorithm executed in step S32. The parameter value (when the adjustment target parameter is plural, the combination of the parameter values used in the defect detection algorithm) is determined to be an inappropriate value (step S36). In other words, the target parameter value group created in the defect detection algorithm executed in step S32 is determined as an inappropriate parameter value group. 124049.doc -33- 200831887 Then, it is judged whether or not the value set for each of the .If ^ ^ ^ ^ positive object parameters (in other words, the value set for each adjustment object parameter in the w疋 object parameter value group) is The maximum value of the adjustment width of the parameter (step S37). When it is determined that there is at least one adjustment target parameter which is not set to the maximum value of the * width, the parameter value (i.e., the parameter value constituting the determination target parameter value group) is changed (step S38). More specifically, this processing is performed in the following manner. When the m object parameter is i, the current setting value of the adjustment target parameter is changed to a value in which the degree of division of the adjustment target parameter is increased, and is obtained as a new determination target parameter value group. . :t right will set the value set by the adjustment object parameter to the maximum value of the adjustment i degree of the parameter'. At the time point when the parameter value completes the defect detection algorithm, it is judged that the appropriate value for all trial values has been completed. If it is judged (YES in step S37), the parameter setting change is not performed, and the process proceeds to step S39. When the adjustment target parameter is two types, the state of one of the fixed adjustment target parameters (set to the second adjustment target parameter) is maintained, and the other adjustment target parameter is set to the first adjustment target parameter. The currently set value 'setting is changed to a value that increases the degree of division of the adjustment target parameter' and is acquired as a new determination target parameter value group. When the value set for the first adjustment target parameter is the maximum value of the adjustment width of the adjustment target parameter, the value set currently set for the second adjustment target parameter is changed to the number of divisions of the adjustment target parameter. The value of the degree, and the first adjustment target parameter value is set to the adjustment again. 124049.doc • 34- 200831887 The adjustment width of the object parameter f π 'minimum value' is obtained as a new judgment target parameter value group. -: Right, the value set for each of the first and second adjustment target parameters is set to & is the maximum value of the adjustment of the U number, and then the time point at which the defect detection algorithm is completed is judged as The test of adjusting the parameters of the object has been completed. If it is judged whether it is appropriate or not (that is, YES in step S37), the parameter setting change is not performed, and the process proceeds to step _9. When the number of σ and sub-images is three or more, the parameter of the adjustment target is two types. The situation is changed. The parameter value is changed in order until the adjustment value of the parameter to be adjusted is determined to be appropriate for all combinations. Referring again to Figure 1〇. In step S37, it is determined that the value set for the majority of all adjustment targets is the adjustment width to the maximum value of the parameter (ie, the trial value of the adjustment target parameter has been added to all combinations (ie, the μ value can be obtained) Group) In the case of performing a defect detection algorithm), then the parameter: decision unit m determines the adjustment value of the adjustment target parameter. More specifically, the processing is performed in the following manner. The "hundredness" judges whether or not there is a parameter value judged to be an appropriate value (in the case where the adjusted logarithm is plural, the group of parameter values judged to be an appropriate value =, S39). More specifically, it is judged whether or not the parameter value in which the d is broken to an appropriate value is stored in the storage unit 12. When the parameter value determined to be an appropriate value is not stored in the memory unit 12, the 翏 value appropriateness determining unit 171 notifies the operator of the fact that the appropriate parameter value is not found (step S40). For example, the warning of "parameter adjustment failure" is displayed on the display unit 14. 124049.doc -35- 200831887 When the value of the parameter judged to be appropriate is appropriate or not, the parameter parameter is plural = Γ binary value, the optimal parameter value (in adjustment The value (step is called. More scorpion and the second is to take the combination of the good parameter values) is selected as the median value selected as the adjustment value L' will be judged as the appropriate value of the trial value under the heart of the way to do the processing. ::'If one (four) chooses the trial value of the adjustment object parameter from the trial value (four) to the appropriate value

Ta〇 and decided to adjust the value. When the number of adjustment target parameters is two, as shown in FIG. 11(b), the set of the parameter value group selected by the adjustment target parameter and the trial value group of the second adjustment target parameter is selected. It seems to be the median Tao and is the adjustment value. The case where the adjustment target parameter is three or more types is the same as the case where the adjustment target parameter is two types, and the median of the set of the combination of the parameter values determined to be the appropriate s value among the combinations of the trial value of the adjustment target parameter is selected. And decided to adjust the value. Thereby, the processing of step S5 is ended. &lt;3. Effect&gt; In the defect inspection device according to the first embodiment of the present invention, when the parameter value related to the image processing is adjusted, the target data creation unit 140 receives the operator's instruction to correct the result data D2. The target data D3 is created, and the parameter adjustment processing unit 170 determines the parameter value of the result data 〇2 that matches the target data D3 as the adjustment value. Thereby, even if the operator does not directly set the input parameter value (that is, even if the trial and error method is used to tentatively set the desired 124049.doc • 36 - 200831887 parameter value), the correction information can be provided only for the result data D2. The desired parameter values are easily obtained. Further, the parameter adjustment processing unit 17 determines the adjustment value as the value of the result data D2 which is obtained in accordance with the target data D3, whereby the parameter adjustment can be accurately performed. Moreover, appropriate parameter adjustments can be easily made, so that parameter adjustments that were previously performed in the manner of a technician service task can be changed to user tasks. This can reduce maintenance costs. In particular, in the defect inspection apparatus 1, the parameters Ρ1 to Ρ4 for determining the detection sensitivity of the defect can be easily and accurately adjusted, so that the defect inspection device 1 can detect the defect by the detection sensitivity desired by the operator. In particular, the target data creation unit 140 displays the inspection target data 01 on the display unit 14, and receives an input of the correction instruction for the result data 自2 from the display surface, so that the operator can easily and accurately input the correction instruction for the result data D2. . As a result, parameter adjustment can be performed more simply and accurately. The eighth parameter value appropriateness determining unit 1 71 determines whether the parameter value is an appropriate value while changing the parameter value of the adjustment target (more specifically I, by judging the result data obtained by each parameter value 02 Whether it is consistent with the target material D3 and determining whether the parameter value is an appropriate value), the parameter value determination unit 172 determines the adjustment value of the adjustment object parameter to an appropriate value, so that the tool can be confirmed; The result of consistent data is the spring value. The multi-eight-eight parameter value determining unit 1 72 selects an episode consisting of parameter values determined to be appropriate values. The median will be determined as the adjustment value of the adjustment target parameter, and the best value of the appropriate value can be obtained (ie, the result data obtained with the target resource 124049.doc -37-200831887 material D3 is obtained most accurately). The parameter value of D2 is determined as the adjustment value of the parameter. In particular, the adjustment constant determining unit 160 adjusts the adjustment width of the object parameter according to the input of the operator, so that the adjustment range of the parameter can be narrowed. As a result, parameter adjustments can be made quickly. Further, the adjustment constant determining unit 160 adjusts the number of divisions of the object parameters in accordance with the operator's input specification, so that the parameter values obtainable in the adjustment range can be reduced. Thereby, the parameter adjustment can be performed more quickly. In particular, the adjustment target parameter specifying unit 150 specifies the parameter of the adjustment target from among the parameters related to the defect detection image processing based on the input of the operator. This allows parameter adjustments to be made quickly. [Embodiment 2] <1> Configuration> The defect inspection apparatus which is one aspect of the data processing apparatus according to the second embodiment of the present invention is configured by the same apparatus as the defect inspection apparatus 1 described in the first embodiment. Implementation 'The description is omitted. When the components of the defect inspection device of the second embodiment are shown in the following, the same components as those of the defect inspection device 1 of the first embodiment are denoted by the reference numerals used in the description of the first embodiment. &lt;1-1·Configuration relating to parameter adjustment&gt; The defect inspection apparatus of this embodiment has a function of adjusting parameters of image processing of the pattern inspection/printed matter inspection. Fig. 12 is a view showing the configuration of the parameter adjustment function. The defect inspection device of the embodiment includes the adjustment target parameter determination unit 124049.doc -38 - 200831887 250 instead of the adjustment target positive object parameter specifying unit 150, and is different from the state defect inspection device. : Guangming Ming. In addition, the description of the same aspects is omitted, and the same components are denoted by the same reference numerals.

The homology image parameter determination unit 250' and the adjustment target parameter specifying unit 15 are different in the parameters Ρ1 to Ρ4 of the phase defect detection image processing, and the specific adjustment target misalignment 'image parameter determination unit 25' is not specified by accepting the operation k. The object parameters are adjusted, but are made based on the ticket data. The operator who is connected to p 14G indicates the correction target parameter for the correction of the result data. The adjustment target parameter determination unit 25A includes a feature amount acquisition unit 251, a corresponding parameter determination unit 252, and an adjustment target parameter determination unit 253. The feature amount acquisition unit 251 obtains a specific feature amount from the operator's acceptance instruction for the result feed D2 (hereinafter simply referred to as "correction instruction"), which is received by the target data creation unit 14A. μ here, with reference to FIG. 13 to FIG. 15 , a more specific aspect of the feature, wherein the result data D2 is illustrated in FIG. 13( a ), and the operator is illustrated in FIG. 13 ( b ) a) The target data d3 produced by the correction instructions E1 to E3 of the result data D2 exemplified in a). Here, it is assumed that the correction instruction E1 is an instruction to correct an isolated point (differential area VI) located in the correction target area (1) from the appropriate area to the defective area, and the 'no correction indication E2' is a plurality of pieces to be located in the correction target area U2. The isolated points (differential regions V2a, V2b, V2c, V2d) are corrected from the appropriate regions to an indication of the defective regions. Further, it is assumed that the correction instruction e3 is an instruction to correct the difference area 124 〇 49. doc - 39 - 2008 31 887 V3 from the appropriate area to the defective area. The feature quantity of the correction instruction is a value for determining a parameter element to be adjusted in order to create a result data 相应2 corresponding to the correction instruction content. The feature amount acquisition unit 251 acquires three pieces of information of the area S, the number A, and the brightness distribution 作为 as feature amounts based on the respective correction instructions. The area s is the total area of the difference area ν associated with a correction indication. For example, the area s obtained from the correction instruction Ε2 is the sum of the areas of the different difference areas V2a to V2k included in the correction target area υ2. Further, for example, the area s obtained based on the correction instruction E3 becomes the area of the difference area ¥3. The number A is the number of difference regions v associated with a correction indication. For example, when there are four independent difference regions VU to V2d in the correction target region m as indicated by the correction instruction E2, the number a is "4". On the other hand, as shown in the correction instruction E1, when there is only one independent difference region V1 in the correction target region (1), the number A is .... Again, such as repair

U is just not E3. $' is the case of directly specifying the difference area ν as "1". The luminance distribution Η is a luminance distribution of a specific region of the inspection target data m (i.e., a region in which V is in the self-difference region, and U in the V starts to occupy a specific range, hereinafter referred to as "luminance amount acquisition region W"). &gt; Here, the acquisition of the luminance distribution is more specifically described. In order to obtain the knives, the biliary object parameter determining unit (4) first obtains the region w. More specifically, first, the specific difference zone is taken. (In the case where there are a plurality of difference regions V associated with the correction indication, the center of all the difference regions V), the center V is specified. For the center of the light 124049.doc 200831887 Measure the acquisition area w. Then, the adjustment target parameter determination unit 25 obtains the luminance distribution of the luminance amount acquisition area Jia of the inspection target data (1). For example, when the situation of the indication ei is corrected, the center of the luminance amount acquisition region W1 coincides with the center v〇 of the difference region νι (refer to Fig. 14(a)). Here, in the inspection target data (1), as shown in Fig. ( (4), the different region... is a point existing in the white background, and the luminance distribution of the luminance amount acquisition region w is obtained, for example, as shown in Fig. 15 (4). Figure. Further, for example, when the indication Ε3 is corrected, the center of the luminance amount acquisition region w coincides with the center Vo of the difference region V3 (see Fig. 14 (1?)). Here, when the difference region V3 is located in the inspection target data 〇1 as shown in FIG. 14(b) at the boundary portion between the pattern and the background, as the luminance distribution of the luminance amount acquisition region 1, for example, as shown in FIG. 15 (1) )) The column chart shown. On the other hand, the difference area V3 is in the inspection target data (1), as shown in Fig. Η (where the 灰 is located in the region of the gray-scale change region, the brightness distribution of the region W is obtained, for example, as shown). Referring again to Fig. 12 . The corresponding parameter determination unit 252 determines the type of the parameter to be adjusted based on each feature amount acquired by the feature amount acquisition unit 251. Here, the determination method will be more specifically described. The determination based on the area S is performed in the following manner. First, it is judged whether the value of the area s is larger than a specific value si. Here, when it is judged that the value of the area 8 is larger than the specific value, it is judged that the comparison parameter p 3 should be adjusted. When it is judged that the value of the area S is not larger than the specific value 81, it is judged whether or not the value of the area S is larger than the specific value s2 (where sl> s2). Here, 124049.doc -41 - 200831887 determines that the value of the area s is greater than a specific value to adjust the parameter P1.疋 is 6 weeks. On the other hand, when it is judged that the value of the area S is not larger than the value s2, it is determined that the isolated point parameter P4 should be adjusted.

The determination based on the number A is performed in the following manner. First, it is judged whether the value of the number A is larger than a specific value nl. Here, the swaying parameter is determined to be the swaying parameter when the number A is greater than the specific value of the heart. On the other hand, when the number A is not larger than the specific value nk, it is determined that the isolated number P4 should be adjusted. &gt; The sigh based on the luminance distribution is performed as follows. First, the adjustment target γ determination unit 250 determines that the histogram shape of the acquired luminance information is the brother 1 distribution shape hl", the "second distribution shape h2", and " The third distribution shape h3" is the tenth. The result is that when it is determined that the shape of the histogram is the shape of the ^^ shape ^, it is determined that the isolated point removal parameter P4 should be adjusted to determine that the shape of the histogram is the second distribution (4). When the fuzzy number P1' is adjusted to determine that the shape of the histogram is the third distribution shape, it is determined that the comparison parameter P3 should be adjusted. Here, a method of determining which of the distribution shapes hi to h3 in the shape of the histogram will be described with reference to Figs. 15 and 16 . Here, Fig. 16 is a flow chart showing the type of the shape of the column chart. First, the value (maximum value m) of the maximum luminance in the histogram distribution is obtained (step S101). Then, the reference value N is calculated (step S102). The reference value N is the value of one-half of the maximum value (N = M/2). 124049.doc -42- 200831887::, determine whether the distribution width when the brightness value is the reference value N is G (SH) 3). Here, as shown in Fig. 15 (b), when the luminance value enters the value N, two or more independent '··, the distribution width of the soil is larger than a specific value. Ge Zhi ^ when 'judge the widest step::. In the case where it is judged that the distribution width is larger than the specific value and the distribution width is larger than the specific width, it is determined that the core knives are the third distribution shape! (Step s1 〇 4). In the case of the step S1〇3, it is judged that the distribution width is not larger than (i.e., there is no case where the distribution width is larger than the specific width), and two or more distribution regions are present at the following: (step S105). When the two or more distributed shapes appear in the break, the 疋 &gt; map is distributed as the second distributed shape h2 (step si 〇 6). In step (4), it is judged that two or more distribution regions do not appear. The column profile distribution is the first distribution shape h1 (step S107). ^ The above-described determination processing is performed on the shape of the histograms of the graphs (4) to (4) = the graph of the de = 15 (4) is the first distribution shape... the adjustment should be performed = the parameter P4 is removed... the column diagram of the judgment _ (8) is = two fuzzy parameters Pl . Further, it is determined that the histogram of Fig. 15(C) is the shape 3 of the knife 3, and the comparison parameter p3 should be adjusted. ~, people' as shown in Figure 12. The adjustment target parameter determination unit 253 counts the number of determinations of the last name outputted by the corresponding parameter determination unit 252. More... Synthesis, the final specific adjustment object ^ More /, body and Lu, calculate the storage decision results in the proportion of each parameter in the order of the two parameters are determined as the parameters. τ 124049.doc -43- 200831887 &lt;2. Processing&gt; The parameter adjustment processing of the defect inspection device 1 of the second embodiment will be described. The overall process of parameter adjustment processing is shown in Figure 5. In this embodiment, the specific processing of the adjustment target parameter is performed as follows (that is, the processing corresponding to step S4). &lt;2-1 - Specific processing of adjusting object parameters &gt;

(Fig. 17 is a flowchart showing the specific processing of the adjustment target parameter executed in the embodiment. First, the feature amount acquisition unit 25 1 corrects the result data D2 accepted by the target data creation unit 140 in step S3, respectively. Instructed to obtain a specific feature quantity (ie, area s, number A, and brightness sub-step Slu). For example, when the target data D3 is created, as shown in FIG. 13, when three correction instructions E1 to E3 are accepted, the feature The quantity acquisition unit 251 acquires the area S, the number a, and the brightness distribution η for each of the correction instructions E1 to E3. Then, the corresponding parameter determination unit 252 determines the type of the parameter to be adjusted based on each feature quantity acquired in step S111 (step Further, the determination result is sequentially transmitted from the corresponding parameter determination unit 252 to the adjustment target parameter determination unit 253, and the adjustment target parameter determination unit 253 stores the determination result. All the feature quantities obtained by the corresponding parameter determination unit 252 are completed. When it is determined that the adjustment target parameter determination unit 253 has obtained the determination result for all the feature amounts (YES in step S113), the adjustment target parameter determines the $section&quot;253 The result of the determination is integrated and specified as the parameter of the adjustment target (step SU4). More specifically, the proportion of each parameter in the determination result of the storage is calculated, and the two parameters are determined to be 124049 in descending order. In the step (4), the feature amount acquisition unit 251 acquires the area S, the number A, and the brightness distribution 对 for each of the correction indications, and the corresponding parameter determination unit in step SH2. 252, the types of the parameters to be adjusted are sequentially determined based on the obtained nine feature amounts, and the result of the determination is transmitted to the tuning target parameter determining unit 253, and the adjustment target parameter determining unit 253 stores the nine determination results as shown in Fig. 18 . In the example of FIG. 18, the adjustment target parameter determination unit 253 refers to the isolated point removal parameter Ρ4 which is the parameter having the largest proportion among the stored determination results, and the fuzzy parameter whose ratio is the second largest is called the modulating parameter. 17. After the two adjustment target parameters are determined in step S114, the adjustment target parameter determination unit 25 inputs the setting screen for inputting the parameter adjustment (see FIG. 9). The display unit i4 is displayed on the display unit i4 (step si 15). Further, in the embodiment, the area h11 in which the face is received is not the area for receiving the selection input of the parameter of the object of the object, but is used as the display adjustment target. In this case, the adjustment target parameter determination unit 250 displays a check state of the check box of the parameter determined as the adjustment target parameter in the check box displayed in the region hi. The area h2 and the area h3 are the same as in the first embodiment. Then, the adjustment target parameter determination unit 250 receives the adjustment width and the number of divisions of each adjustment target parameter (step S116). This processing is the same as the processing of step S23 (Fig. 8) described earlier. Then, when the input of the adjustment width and the number of divisions of each adjustment target parameter is completed, and the operation of the decision panel is accepted (YES in step S117) 124049.doc •45-200831887, the adjustment target parameter determination unit 250 reads in The adjustment width and the number of divisions determined for each adjustment target parameter (step S118). This ends the specific processing of the adjustment target number. ^3·Effects> In the defect inspection device according to the second embodiment of the present invention, the cookware and adjustment target parameter determination unit 250 detects the image from the defect based on the operator's _$社_^ and the positive deduction. The parameters of the processing are determined to be the parameters of the adjustment pair, and therefore, the types of parameters to be adjusted can be appropriately selected in order to obtain the result data with the target data D3. Thereby, parameter adjustment can be performed quickly and accurately. [Third Embodiment] &lt;1. Configuration&gt; The defect inspection device which is one aspect of the data processing device according to the third embodiment of the present invention is constituted by the same device as the defect inspection device 1 described in the third embodiment. The implementation is omitted, and the description thereof is omitted. In the following, when the components of the defect inspection device of the jth embodiment are shown below, the reference symbols used in the description of the embodiment of L) are used. The defect inspection device of this embodiment is the same as the defect inspection of the second embodiment, and includes an adjustment constant determining unit 丨7〇. However, in this embodiment, the adjustment constant determining unit 170 does not determine the adjustment value by selecting the median number of the clusters formed by the appropriate values, but first determines the appropriate value to be detected as the adjustment value. &lt;2. Processing&gt; The parameter adjustment processing of the defect inspection device 第 according to the third embodiment will be described. The overall process of parameter adjustment processing is shown in Figure 5. Among them, in the implementation of 124049.doc • 46-200831887, the determination of the adjustment value of the parameter is performed in the following manner (that is, equivalent to the processing of step S5). &lt;2-1_ Determination of Adjustment Values> Fig. 19 is a flowchart showing the process of determining the adjustment value of the adjustment target parameter executed in the embodiment. First, the same processing (steps S211 to S214) as the steps S31 to S34 (Fig. 1A) described herein will be performed. In step S214, when it is determined that the defective area portions of the two image data are identical, the parameter value appropriateness determining unit 171 sets the parameter value used in the defect detecting algorithm executed in step 8212 (the adjustment target parameter is plural) In the case of a case, a combination of parameter values used in the defect detection algorithm is judged to be an appropriate value. In this case, the parameter value determined to be the appropriate value is determined as the adjustment value of the adjustment target parameter (step S215). That is, as described in the first embodiment, it is not necessary to perform a defect detection algorithm for all combinations of the trial values of the respective adjustment target parameters, and the parameter value can be determined as the adjustment value when an appropriate parameter value is found. That is, first, the detected appropriate value is determined as the adjustment value. On the other hand, in step S214, when it is determined that the defective area portions of the two image data do not coincide, the parameter value appropriateness determining unit 171 sets the parameter value used in the defect detecting algorithm executed in step S212 (in When the object parameter is adjusted to be plural, the combination of the parameter values used in the defect detection algorithm is determined to be an inappropriate value (step S216). In this case, it is then determined whether the value set for each adjustment target parameter is the maximum value of the adjustment width of the parameter (step S217). 124049.doc -47· 200831887 In the case where it is determined that there is at least one adjustment target parameter which is not set to the maximum value of the adjustment width of the parameter in step S217, the setting of the parameter value is changed (step coffee). This processing is the same as the processing of the above-described well S38 (Fig. 1A). In the case of eve fv 2 1 7 , the value set for all the adjustment artifact parameters is determined as the maximum value of the adjustment width of the parameter (ie, the defect detection algorithm is implemented for all combinations of the martial values of the adjustment target parameters). In the case of the case, the parameter value appropriateness determining unit 171 passes the subject matter in which the appropriate parameter value is not found to the operator (step S219). That is, even if all of the trial/depreciation values of the respective adjustment target parameters, and the σ Λ defect detection algorithm, when the appropriate parameter value is not found, the processing of step S219 is performed. This completes the decision processing of the adjustment value of the adjustment target parameter. <3>Effects> In the defect inspection device according to the third embodiment of the present invention, the parameter value of the adjustment target parameter is determined as the adjustment value by first determining the appropriate value, so that the parameter adjustment can be quickly performed. . [Fourth Embodiment] &lt;1. Configuration&gt; The defect inspection device of the data processing device according to the fourth embodiment of the present invention is constituted by the same device as the defect inspection device described in the first embodiment. The implementation is omitted, and the description thereof is omitted. In the following, when the components of the defect inspection device of the second embodiment are shown, the reference symbols used in the description of the first embodiment are used. &lt;1-1·Configuration relating to parameter adjustment&gt; 124049.doc -48· 200831887 This embodiment of the defect inspection apparatus has a function of adjusting image processing parameters related to the pattern/printed matter inspection. Figure 20 is a diagram illustrating the composition of the parameter adjustment function. In the defect inspection apparatus of this embodiment, the state in which the parameter adjustment value is determined is different from that of the defect inspection device 1 of the first embodiment. Hereinafter, only the differences from the defect inspection device 1 of the first embodiment will be described, and the same points will be omitted. Further, the same components are denoted by the same reference numerals. The defect inspection apparatus of this embodiment has a function of adjusting image processing parameters related to the pattern/printed matter inspection. Fig. 20 is a view for explaining the configuration relating to the parameter adjustment function. In addition, the following is referred to FIG. 2 and FIG. 22 as appropriate. Figure 2 1 and Figure 22 illustrate the conceptual diagrams of the adjustment values using the genetic algorithm. The defect inspection device of this embodiment includes a parameter adjustment processing unit 47 (). The parameter adjustment processing unit 470 determines the adjustment value of the adjustment target parameter using the genetic algorithm. More specifically, the genetic algorithm is used to extract a parameter value group that provides the result data D2 from the target data D3 from a plurality of parameter value groups corresponding to all combinations of the adjustment target parameters (the trial values). Further, the value of the parameter value group constituting the captured parameter is determined as the adjustment value of the adjustment target parameter. As shown in FIG. 20, the parameter adjustment processing unit 470 includes the first generation individual group generation unit 471, the fitness value acquisition unit 472, and the pair. The generation unit 473, the intersection processing unit 474, the mutation execution unit 475, and the optimal parameter value determination unit 476. The first generation individual group generation unit 471 generates a specific number n (where N is a specific natural number), and the value can be arbitrary. In addition, in the example of Fig. 21, n=4) individual 124049.doc -49-200831887 group and obtained as the first generation of the individual group. Each body has a chromosome composed of a single base. More specifically, the gene refers to any value in the range of 0·0 to 1.0, and the values of the genes of the first generation of each body are randomly selected. Also, &amp; $ ^ ^ A ^ The gene contained in the chromoplast The number M (in the case of Figure 2丨 'Μ) and defect detection Like the number of parameters of the processing, as shown in Fig. 22, each of the genes 1, 2, ... 7 corresponds to any one of the parameters related to the defect detection image processing. That is, each chromosome corresponds to the parameter value group. One or more parameters selected from the parameters of the defect detection image processing are set as the adjustment target parameters, so that each gene corresponds to any one of the adjustment target parameters. That is, in this case, the number of genes The fitness value acquisition unit 472 obtains an individual's fitness value. The fitness value acquisition unit 472 performs the following processing on each of the N individuals constituting each generation to obtain individual adaptations. In addition, the acquired adaptive value is stored in the memory unit 12 in accordance with each body. The process of obtaining the adaptive value will be described with reference to Fig. 22. When the adaptive value is obtained, first, the individual included in the individual is explained. One (in the example of Fig. 22, Μ=7) gene obtains a parameter value. That is, a parameter value group is obtained from a chromosome. The process of interpreting the gene to obtain the parameter value is performed by the following (Formula 1) Show In the case of (丨), the value of the "Gi" soil mouth 'Pi' is the value of the parameter corresponding to the gene (丨=1, 2,···, Μ). "min" and rmax" are the minimum and maximum values of the adjustment width set for this parameter. Also, "div" is the number of divisions set for this parameter. Also, rint { }" means discarding the decimal point. The following values are processed by 124049.doc -50- 200831887.

Pi=min+(max-min)*int{Gi*div}/div·. (Expression 1) Then, the defect detection algorithm is executed on the obtained parameter value group, and the result data D3 is obtained. Further, the obtained result data D2 is compared with the target data D3, and the individual fitness value is calculated. Here, the "adaptive value" refers to the area ratio of the portion of the result data D2 that coincides with the target image data D3 (the ratio of the area of the coincident portion to the entire area of the image). That is, the fitness value of the individual indicates that the result data D2 corresponding to the parameter value group of the individual (chromosome) is consistent with the target data D3, and the larger the value, the better the individual (providing the height of the target data D3) The result of the consistent result data D2 corresponds to the individual of the parameter group). Among them, the fitness value can take the f value of the range of 〇~丨. For example, it can be said that the individual with the fitness value of "1" is the best individual corresponding to the parameter value group of D3 which is the full-form result of the target data (4). The generating unit 473 generates [from the n individuals of the same generation [(n/uj group (where N is an even number. When N is an odd number, it is a group). More specifically, t 'first, since the same generation ^ In the individual groups, according to the principle of m from the adaptation value, (N_2) individuals (N-1 in the case of N odd numbers) are selected. The individuals selected here become the mothers of the next-generation individuals. The so-called natural elimination principle refers to the principle that the superior individual (that is, the individual with high fitness value) is inherited from the next generation to the generation, and the probability that the better individual among the N individuals of the same generation is selected as the mother is more and more For example, the fitness value is set to the probability of selection as the parent. Alternatively, it may be configured as follows: the individuals are ranked in the order of small to large, 124049.doc -51 - 200831887, self-proclaimed The order individuals assign the selection probability of the higher value in order. The generation unit 473 selects (N-1) (9) which are odd according to the above principle, and (N-1) are different from each other. Individuals are set to pairs, resulting in [(^72)-1] group (when 1^ is odd) (1 ^ 1) / 2 group) pair. Furthermore, by making the following crossovers and mutations, the next generation of individuals is generated, and the number of individuals is adjusted so that the number of generations of the next generation is the same as that of the previous generation (ie, even after generations of individuals) The number of individuals used by the father in the above-mentioned f-month shape (that is, the number of individuals generated by the intersection as described below) is (N_2) when the number is even, and when n is an odd number. It is (ND, and if so, the number of individuals generated in one generation is reduced. Thus, the adjustment processing to supplement the reduction is performed.

When Yu is an even number, reduce two individuals. In this case, for example, an individual is supplemented with a refined strategy (a method of directly retaining individuals with high fitness values as individuals of the next generation), and is supplemented with an individual with a randomly selected gene. On the other hand, in the case of (10) odd numbers, one individual is reduced. In this case, for example, the elite strategy can be supplemented by one or more. The number of pairs generated here can also be (Ν/2) group (where Ν ^ even number is used. When Ν is odd, it is [( Ν+1)/2] group). In this case, the number of individuals used in the transaction is Ν when ν is even, and ' is '(Ν+1) when Ν is odd, so it is generated when Ν is odd - The number of generations has increased. Therefore, in this case, when the Ν is an odd number, after the cross processing, an individual is randomly selected and deleted from the generated generations 124049.doc -52-200831887, and the number of individuals is adjusted. The intersection processing unit 474 causes the two pairs of individuals to be handed over to produce the next generation of individuals. More specifically, as shown in Fig. 21, a new individual is generated by replacing the gene located closer to the second half at a more specific intersection position Q (where the intersection position is set to a position randomly selected from the chromosome). In this way, two next generation individuals are created by each pair. As shown in Fig. 21, it is very likely that the fitness value of the newly generated individual by the intersection of the superior maternal selected by the previous generation is better than the maternal fitness value. That is to say, it is possible that the newspaper will advance with the generations, naturally eliminate individuals with low fitness values, and produce excellent individuals with high fitness values. Here, the gene shown by hatching in Fig. 21 indicates a gene corresponding to an adaptive value. Here, the case where the individual 2 having the fitness value "〇5" is generated in the second generation is exemplified. Further, as described above, the number of individuals is adjusted here so that the number of generations of the next generation is the same as the number of individuals of the previous generation. That is, when the number of individuals used for the intersection is (Ν-2), the number of individuals generated is N. '(4) The British strategy and the selection of the age are called a fixed reduction. individual. Moreover, when the number of individuals used in the intersection is a situation, a reduced individual is supplemented by an elite strategy. Further, in the case where the number of individuals used in the transaction is (Ν+1), an individual is randomly selected and deleted from the next generation of the generated individuals. The mutation executing unit 475 mutates the generated new individual group gene at a specific probability when a new individual group is generated by the intersection. The term "mutation" refers to the genetic value of each individual resulting from a change in regeneration. By mutating the gene at a specific probability (i.e., randomly changing the gene value at a specific probability), the deviation of the gene = 124049.doc -53 - 200831887 is eliminated, thereby avoiding the risk of falling into a local minimum in the optimization process. The optimum parameter value determining unit 476 determines an adjustment value (a value to be output as an adjustment result) of the adjustment target parameter. More specifically, after an individual of a specific generation (e.g., the Tth generation (where τ is an arbitrary number of 2 or more natural numbers)) is generated, the individual having the highest fitness value among the generations is taken as the best individual. Further, it is judged that the parameter value group corresponding to the best individual chromosome obtained is the optimal parameter value group, and the parameter values constituting the optimal parameter value group are determined as the adjustment values of the respective parameters. In other words, the parameter values obtained by interpreting the genes contained in the best individual chromosome are determined as the adjustment values of the respective parameters. Furthermore, it is also possible to adopt a configuration in which, regardless of the number of generations, the individual is taken as the best individual at a time point of producing an individual having an adaptation value higher than a specific value. &lt;2. Processing&gt; The parameter adjustment processing of the defect inspection device of the fourth embodiment will be described. The overall process of parameter adjustment processing is shown in Figure 5. Here, in this embodiment, the determination processing of the adjustment value is performed as follows (that is, it corresponds to the processing of step S5). &lt;2-1. Decision Process of Adjustment Value&gt; Fig. 23 is a flowchart showing the process of determining the adjustment value executed in the embodiment. In addition, FIG. 2A and FIG. 22 are also referred to below as appropriate. After the processing of step S4 (Fig. 5) is completed, the acceptance screen (10) is again displayed on the display unit Μ. After receiving the operation of the "recalculation" panel in the face G, the process of step S5 is started. 124049.doc -54- 200831887 First, the third generation generation group generation unit 471 generates ^^ individuals constituting the third generation (in the example of Fig. 21, the individual is a step (3) ten here, each generated As described above, the individual has a chromosome composed of M genes that are randomly selected. Then, the fitness value acquisition unit 472 obtains an adaptive value of each of the N individuals constituting the i-th generation, and memorizes the acquired fitness value corresponding to each individual. In the memory unit 12 (step S312), the generation unit 473 generates a group of [(^2)4] from n individuals of the first generation (when N is an odd number, it is a pair (step S313). More specifically In the N generations of the first generation, according to the principle of natural elimination using the fitness value, (N-2) (when N is an odd number, (Ν_υ)) individuals are selected to make the selected ( Ν-2) (when the odd-numbered case is (Ν-1)) Two individuals different from each other are paired, resulting in the [(Ν/2)_ι] group (when ν is odd) , (Nl)/2 group). Then, the cross processing unit 474 causes the pair of pairs generated in step S3 13 to be paired The individual crosses to generate a new individual (step S314). More specifically, as shown in Fig. 21, the gene is replaced by a gene located closer to the second half than a specific intersection position randomly selected from the chromosome, and newly generated (Ν_2) In the case of an odd number, it is (N_i) individuals. Furthermore, the number of individuals is adjusted here so that the number of generations of the next generation is the same as that of the previous generation. That is, the number of individuals produced is In the case of (N_2), use an elite strategy to supplement an individual and supplement an individual with a randomly selected gene. On the other hand, use the elite strategy when the number of individuals produced is (N_丨) An individual is added. 124049.doc -55- 200831887 In turn, the mutation executing unit 475 generates a gene mutation in the new individual generated in step 8314 with a specific probability (step S3 15). Newly generated by the processing of steps S314 to S315 1 individual, the N individuals (in the case of Fig. 21, the individual Γ, 2, 3, 4, 4) are the individual groups constituting the second generation. Then, the optimal parameter value determining unit 476 Judge Whether or not an individual having a specific generation (the third generation (where τ is an arbitrary natural number)) is generated (step S3 16). In other words, it is judged whether or not the processing of steps S312 to S315 is performed for a certain number of times π" times) When it is determined in step S3 16 that the τ generation is not generated, the process returns to step S3 12. For example, in the case of τ=3, if the second generation individual group is generated in the process of step S3 15, Then, the process returns to step S3 i 2 again. In this case, the process of step S3 12 to step S3 15 is performed on the second generation individual group to generate the individual group constituting the third generation. When it is determined in step S3 16 that the τ generation is generated, the fitness value acquisition unit 472 acquires the fitness values of the N individuals constituting the Tth generation, and the optimal parameter value determination unit 476 has the τ generation The individual with the highest fitness value is the best individual (step S3 17). For example, in the case of T = 3, if the third generation individual group is generated in the processing of step S3 15, the optimal parameter value decision portion 476 manipulates the individual having the highest fitness value among the third generation individuals as the best individual. Further, the determination processing of step S3 16 may be changed, and after the processing of step S3 12 (i.e., after the adaptation values of the respective bodies are acquired), processing for determining whether or not there is an individual having an adaptation value of a certain value is performed. In this case, it is judged that 124049.doc -56- 200831887 has such an individual time point, and the individual is taken as the best individual. And taking the better parameter value determining unit 476, and interpreting the genes included in the best individual chromosome extracted in step S317 to obtain one parameter value, the value of the parameter is determined as the adjustment of each parameter. Value (step S318). Thereby, the processing of step S5 (Fig. 5) is ended. Furthermore, in the above, the following configuration is adopted, that is, the next generation system is paired from the individuals selected according to the natural elimination principle using the fitness value, and f' is generated for the parent fork, but the next generation individual The production method is not limited to this. For example, a method of directly retaining individuals with high fitness values as individuals of the next generation (elite strategy) can also be used. Also, a method of randomly generating an individual of the next generation can be employed. &lt;3. Effect&gt; In the defect inspection device according to the fourth embodiment of the present invention, the genetic value method is used to take the adjustment value of the adjustment target parameter, so that the optimum parameter value can be quickly obtained. Further, in the defect inspection apparatus of the embodiment, various parameters are respectively associated with a plurality of genes included in the chromosome (that is, each body is associated; a combination of each parameter value) is adapted to use the nasal expression method. A good individual with a high value (ie, a combination of the best adjustment values for the target data D3). By 'this, even if the type of the adjustment target parameter is increased, the optimal adjustment value can be quickly obtained. [Modifications] &lt;First Modifications&gt; In the above embodiments, the defect detection algorithm is used in a specific order 124049.doc -57· 200831887 to perform blurring processing, shaking processing, comparison processing, and isolated point removal processing. The defect detection algorithm implemented by the defective device 1 is not limited to this aspect. For example, each of the image processing described above may be changed or used in combination to incorporate a core specified blur processing or a full processing into the defect detection algorithm. Moreover, the parameters required for the various processes incorporated can be adjusted using the above-described aspects. That is, in the above-described embodiment, the plurality of parameters of the defect detection image processing include the blur parameter P1, the shake parameter P2, the comparison parameter P3, and the isolated point removal parameter P4', but the parameters of the defect detection image processing are not limited thereto. . For example, the filtering radius value of the filtering process and the color range that should be captured in the processing of the specific color range (for example, the L value of the Lab color space, the range of the a value and the b value), the full width of the full processing, The refinement width of the refinement processing, the filter radius value in the detection processing of the edge component, and the value of the filter 1), and whether or not to perform the mask value of the various processing such as mask synthesis processing or comparison processing, and the area range of the area determination processing ( Various parameters such as the minimum area value and the maximum area value), and the maximum contour length value of the feature amount extraction processing are set as adjustment target parameters. Further, it is not limited to the parameters of the defect detection image processing, and various parameters (for example, the illumination angle of the image scanner 5, the light source color, the optical magnification, the image reading, or the noise removal) are related to the processing of the inspection target data D1. Various adjustment variables, etc.) may also be included in the adjustment object parameters. Further, it is also possible to include whether or not the defect detection algorithm processing (ΟΝ/OFF (on/off) of the algorithm processing function itself) is performed as an adjustment target parameter. &lt;Second Modification&gt; In the above-described respective embodiments, the adjustment of the parameter related to the defect detection image processing is performed on the defect inspection device 1 as the data processing device. Although not limited to defect detection image processing, the above aspect may be applied to various image processing (for example, image deformation processing, noise removal processing, image conversion processing, contrast adjustment processing, etc.). In the device, (10) parameter values related to image processing performed by the entire device. As an example, a case where the above-described aspect is applied to a device (uneven inspection device) for detecting unevenness generated in a liquid crystal pattern will be described. Here, the "unevenness" of the liquid crystal pattern means, for example, a color filter cloth as a liquid crystal display portion, and a uniform concentration of Han (red), G (,, '彔), B (blue) should be formed on the entire surface. The uneven concentration of the color produced in each color. In the unevenness inspection device, a meaningful uneven portion generated in an inspection object such as a trousers is specified by performing a specific unevenness detection process. In the pot, the unevenness detecting process is carried out, for example, in the following manner. First, the image data of the inspection object is analyzed. (4) The object to be inspected is generated: unevenness (for example, by performing concentration analysis, filtering processing, contrast F and processing, high speed Fourier transform _ F ^ τ_〇Γιη, ) and - value processing, etc., are performed by an algorithm that performs the processing. And the unevenness of the '^ is evaluated and the meaning of the detection is uneven: for example, the implementation of the group has a full processing 'fine processing, and the image is unevenly judged 6 area, position (four) argument This processing is performed by comparing the determination determination processing and the like. Here, the detection sensitivity of the uneven operation is determined by the user, and the user is fetched with an appropriate accuracy. In order to detect the unevenness with the appropriate (ie, U-sight) sensitivity, the parameter values related to the processing must be adjusted in advance by 124049.doc -59- 200831887. The unevenness inspection of the text form is installed in the defect detection processing unit 130 (S2, Hiroyuki Department) as a phase. The processing unit performs the following diagram: the image is executed, and the image is processed by the object. The figure "detects the image data in the unevenness detection process... the uneven portion of the image, and produces the characteristic: the data produced in the sputum (ie, the result data D2)...., the uneven portion of H Γ

For other function parts related to parameter adjustment, the same as above (refer to Figure 2), 々ά m data t M4G accepts the author of the result material D2 and produces the content of the finger* in the result = image data (target data D3), and the parameter adjustment processing unit 2 adjusts the adjustment value of the target parameter. Thereby, the parameters related to the unevenness detection/processing can be adjusted. &lt;Third Modification&gt; In the above embodiments, the description of the case where the parameter adjustment relating to image processing is performed may be performed in association with various processes of data other than image data (for example, T vector data). Parameter adjustment. As an example of the case where the parameter adjustment relating to the processing of the vector data is performed, the following case will be described: the above-described aspect is applied to the measurement data of the self-contained noise, and the data of the specific object to be measured is accurately obtained by trial and error. Homework. For example, the measurement data is the film thickness measurement data of the substrate (the data obtained by measuring the film thickness of the substrate having the groove (pattern) formed on the surface by a spectroscopic film thickness meter). Usually, the measurement data contains noise. The film thickness measurement data still contains the influence of noise, and it is difficult to directly read the width or depth of the actual groove 124049.doc -60- 200831887 from the measurement data. Thus, the operation of estimating the width or depth of the groove is performed by using the simulation method as described below. First, vector data specifying the appropriate width and depth (i.e., vector data representing the shape of the groove) is generated, and the vector data is subjected to spectral simulation. In other words, it is assumed that measurement data obtained by measuring the shape of the groove represented by the vector data by a spectroscopic film thickness meter (hereinafter referred to as "hypothetical data") is generated. If the hypothetical data is consistent with the state in which the self-measurement data is removed from the noise, it can be said that the information provided by the hypothesis is free of information to correctly represent the width or depth of the groove. Therefore, the operator uses the trial and error to explore the vector data that provides such hypothesis data (more specifically, after comparing the obtained hypothesis data with the measured data, the shape of the vector data is changed to eliminate different parts generated between the two data. Moreover, while repeating the self-changing vector data, the hypothesis data is generated again, and the obtained hypothesis data is compared with the measurement data to explore. The adjustment function of the above parameters can be applied to the job. In this modification, a processing unit (vector data generation processing 卩) for performing generation processing of vector data is provided as a functional unit corresponding to the defect detection processing unit 图3〇 (Fig. 2). The processing portion generates vector data representing grooves having a specific width and depth. Here, the width ambiguity of the groove expressed is defined by the parameter value supplied to the generation processing unit of the vector data, and in the modification, the parameter is set as the adjustment object parameter. The hypothesis data is obtained by performing specific spectroscopic simulation on the vector data generated by the vector data generation processing unit. In this modification, the hypothetical data obtained by the household is the result data D2. For other function parts related to parameter adjustment, it is the same as ±Representation __ 124049.doc -61 - 200831887 (Refer to (4) ~ that is, 'target f material production department 丨 transfer operator's correction instruction for result data D2 Further, the data (the data #3) in which the instruction content is reflected in the result data D2 is created. In the modification, the operator corrects the result data D2 to remove the noise from the self-measurement data of the created target data. The parameter adjustment processing unit 170 obtains a parameter value that can provide the result data D2 with the target data D3_ as an adjustment value. That is, (4) parameter adjustment

The processing unit m obtains a parameter value which can generate an appropriate vector: (4) (i.e., vector data which can provide a hypothetical data in accordance with the state in which the noise is removed from the measurement data). The vector data generated using this parameter value becomes the data that accurately represents the width or depth of the groove. Thus, the operation of estimating the width or depth of the actual groove based on the measurement data can be easily and accurately performed by using the parameter adjustment function. &lt;Fourth Modification&gt; The above embodiments are also described as a third modification, and are a technique that is effective in the field of simulation. For example, in various article manufacturing steps, the following operations may be performed as follows: (4) The shape of the material is processed by simulation according to the shape of the material before processing, and the shape after the reinforcement is confirmed, and the adjustment function of the above parameter can be applied to the target as the target. operation. In the modification, the processing unit (material shape generation processing unit) that performs the process of generating the shape f material indicating the shape of the material before the garnish is provided as the defect detection processing unit=image extraction function unit. The processing unit generates an I-shaped bead material having a shape of the material to be fixed. Wherein the shape represented by the shape is supplied to the shape of the material 124049.doc • 62- 200831887 The parameter value parameter of the processing part n & 'In this modification, the parameter is set to adjust the object, the object is lost, and the performance is also The coarse-grained state of the mesh of the shape is a shape obtained by processing the shape generated by the processing section of the material shape, "the simulation of the work industry", and obtaining the shape information of the shape of the material. . In this modification, the heartbeat obtained here is the result of the poor material D2. The other functional units related to phase=parameter adjustment are the same as the above-described embodiments: first, Fig. 2). In other words, the target data creation unit 140 accepts an operator's instruction to correct the result and the material D2, and creates a data (target data D3) in which the instruction content is reflected in the result value = 2. In this modification, the operator corrects the result Beckham D2 so that the created target material (1) exhibits the processed shape desired by the operator himself. Further, the parameter adjustment processing unit 170 obtains a parameter value which can provide the result data D2 in accordance with the target data D3 as an adjustment value. That is, a parameter value is obtained which can generate data indicating the shape of the appropriate material (i.e., the shape of the material that can provide the processing shape desired by the operator). Thus, the shape of the material used to obtain the desired processed shape can be easily and accurately estimated by using the parameter adjustment function. &lt;Fifth Modification&gt; The function unit for adjusting the parameters described in the above embodiments may be provided with a configuration (parameter evaluation unit (not shown)) for evaluating the acquired adjustment value.

The parameter evaluation unit evaluates whether or not the adjustment value of the adjustment parameter determined by the parameter adjustment processing unit 丨 70 (Fig. 2) is an appropriate value. For example, the data obtained by the defect detection algorithm (final result data) will be compared with the target data D3, and the area ratio of the two data consensus parts will be calculated and obtained. This value is used as an evaluation value of the adjustment value. That is, it can be said that the higher the evaluation value is, the higher the degree of the data obtained by the obtained adjustment value coincides with the target data D3. The evaluation value obtained may be displayed, for example, on the display unit 14 to notify the operator. Further, when the evaluation value is lower than the specific value, (7) the parameter value may be finally fine-tuned, and the parameter adjustment p processing unit 170 may perform the calculation processing of the adjustment value again. &lt;Other Modifications&gt; In each of the above-described embodiments, the defect regions T1 and T2 are displayed in magenta in the inspection target data m, whereby the defective regions are characteristically displayed, but the defects are characteristically displayed. The aspect of the area is not limited to this. For example, a pattern in which a defective area is displayed more densely in an appropriate area, a pattern in which a defective area is displayed in a blinking manner, and an appropriate area in a specific color (for example, a light gray) may be displayed. Further, in each of the above embodiments, the case where the defect region portion of the result data D2 and the defect region portion of the target data D3 are completely identical, the trial value used in the (4) result data D2 is determined as the appropriate adjustment parameter. Value, but even if the defect area of the result data D2 and the defect area of the target data are incomplete, the two data may be considered to be consistent when the degree of the defect is within the specified tolerance, and the result data D2 will be produced. The trial value used is judged to be an appropriate value of the adjustment target parameter. For example, if the area of the defect area that is not in the two data is less than a specific value, the two data may be considered to be consistent, and the trial value is judged to be an appropriate value of the adjustment object 124049.doc -64 - 200831887. Further, in the above-described respective embodiments, the parameter adjustment processing unit 17 causes the parameter value to be sequentially increased from the minimum value in the adjustment range defined by the adjustment width when the parameter value is changed in the predetermined processing of the adjustment value. (Step S3 1 of Fig. 1), but the aspect in which the parameter value is changed is not limited thereto. For example, the parameter value may be sequentially decreased from the maximum value of the adjustment range, or the parameter value may be changed by increasing or decreasing the number of divisions in the median of the adjustment range. Further, the value set by the adjustment may be changed as an initial value. In the first embodiment, the median value of the trial value determined by the parameter value appropriateness determination unit 171 as an appropriate value is selected as the adjustment value (step S4D in FIG. 10, but the median is determined based on the appropriate value. The aspect is not limited to this. For example, a trial value (a combination of parameter values determined to be appropriate values when the adjustment target parameter is plural) can be displayed on the display unit 14, and the operator selects any An appropriate value. Further, in the second embodiment, two τ - 1 solid shells (area S, number Α, and luminance distribution Η) are obtained as feature quantities, but

Jw is also one of the feature quantities: one is formed as a feature quantity. x, information obtained as a feature quantity

Time::: This. For example, the operator can specify the drag area of the correction target area U f as the feature amount. Further, the adjustment target parameter determination unit 253 summarizes the stored determination result, and the feature of the adjustment target 蚪瞀Φ夂 winter batch is weighted. Ten different proportions of each parameter. Further, in the second embodiment, the defect is detected as a defect, and the target parameter specifying unit 150 included in the heartbeat device of the first embodiment is replaced by 124049.doc. 65 - 200831887 The adjustment target parameter determination unit 25 is, but like the parameter identification unit 15. When adjusting the object parameter 2 == adjusting the situation, ' can be set as follows: operation. In this case, the 丨 丨 j j j j j j j j j j j j j j j j j 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 250 The specificity of the object parameters. In the second embodiment, the adjustment is determined by the emergency target parameter determining unit 253.

The parameter with a high proportion is determined by adjusting the two parameters in order to adjust the number of wheat, but it is also possible to select __ or more parameters according to the proportion from low to high and determine the parameters to be adjusted. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a view showing a configuration of a printing system i of a defect inspection apparatus 1 as an aspect of an image processing apparatus according to a third embodiment of the present invention. Fig. 2 is a block diagram showing the configuration related to the parameter adjustment function. 3(a) and 3(b) are diagrams showing the reference data D0 and the inspection target data D1. Fig. 4 (a) and (b) are diagrams showing the result data D2 and the target data D3. Fig. 5 is a flowchart showing the overall process of the parameter adjustment processing of the defect inspection device 1. Fig. 0 Fig. 6 is a flowchart showing the process of creating the target data D3. Fig. 7 is a view showing an example of the configuration of the acceptance screen G. Fig. 8 is a flow chart showing a specific process of adjusting an object parameter. Fig. 9 is a view showing an example of the configuration of the acceptance screen Η. Fig. 1 is a flow chart showing the process of determining the parameter values. 124049.doc -66- 200831887 Figure 11 (a), (b) is a diagram for explaining the selection of the median value of the trial value determined to be an appropriate value as the adjustment value. Fig. 12 is a view for explaining the configuration related to the parameter adjustment function. 13(a) and (b) are diagrams showing the result data 〇2 and the target data 〇3. 14(a) to (c) are diagrams illustrating a luminance amount acquisition area. 15(a) to (c) are diagrams showing a histogram of the degree distribution. Fig. 16 is a view showing the flow of determining the kind of the shape of the histogram. Fig. 17 is a flowchart showing a specific process of adjusting an object parameter. FIG. 18 is a diagram showing the determination result stored in the adjustment target parameter determination unit 253. Fig. 19 is a flow chart showing the process of determining the parameter values. Fig. 20 is a block diagram showing the configuration related to the parameter adjustment function. Figure 21 is a conceptual diagram illustrating the deterministic aspects of the adjustment values using the genetic algorithm. Figure 22 is a conceptual diagram illustrating the deterministic aspect of the adjustment value using the genetic algorithm. Fig. 23 is a flow chart showing the process of determining the adjustment value. [Description of main component symbols] 1 Defect inspection device 130 Defect detection processing unit 140 Target data creation unit 150 Adjustment target parameter specifying unit 160 Adjustment constant determination unit 170 Parameter adjustment processing unit 124049.doc • 67- 200831887 250 Adjustment target parameter determination unit DO Reference data D1 Inspection target data D2 Result data D3 Target data P program 124049.doc -68-

Claims (1)

200831887 X. Patent application scope: 1 . An image processing apparatus, comprising: an image processing mechanism that performs specific image processing; and a target image producing mechanism that accepts a pair as a result of the image processing a correction instruction of the produced result image, a target image in which the correction instruction is reflected in the result image; and a parameter adjustment mechanism that adjusts a parameter value of the image processing so as to be a result of the image processing The resulting image is identical to the target image described above. 2. The image processing apparatus according to claim 1, wherein the image processing means performs image processing as follows: comparing the reference image with an inspection target image obtained by outputting the reference image, and detecting the inspection target image The defective portion generated in the image is subjected to a result image in which the detected defective portion is displayed in a display form different from the non-defective portion; and the parameter of the image processing defines the detection sensitivity of the defective portion. 3. The image processing apparatus according to claim 1 or 2, wherein the target image creating means displays the result image on the display screen, and receives an instruction to correct the result image from the display surface. Input. 4. The image processing apparatus of claim 3, wherein the meal number adjustment mechanism determines whether the result image obtained under each parameter value is related to the target image 124049.doc 200831887 while changing the value of the parameter. And the adjustment value of the above parameters is determined as the parameter value of the resulting image that is consistent with the upper image. The image processing device of claim 4, wherein the parameter adjustment unit determines the adjustment value of the parameter as a parameter value of a result image that matches the target image: number. 6. The image processing apparatus of claim 4, wherein the parameter adjustment means determines the adjustment value of the parameter as a result of initially matching the target image in the process of changing the value of the parameter. The parameter value at the time of the image. 7. The image processing apparatus of claim 4, further comprising: an adjustment range specific mechanism that specifies a range in which the parameter value is changed based on the input set value. 8. The image processing apparatus of claim 4, further comprising: a variation amount specifying means that specifies an amount of change when the parameter value is changed based on the input set value. 9. The image processing apparatus according to claim 3, wherein the parameter adjustment means selects a combination of the result images assigned to the target image by using a genetic algorithm from a combination of values of the one or more parameters. The adjustment values of the parameters are the values constituting the combination of the extractions. 10. The image processing apparatus of claim 9, further comprising: an adjustment range specifying means that specifies a range in which the parameter value is changed based on the input set value. The image processing device of claim 9, further comprising: a variation specific mechanism that specifies a variation amount when the parameter value changes according to the input value of the input. The image processing device of claim 1 or 2, wherein the image processing device according to claim 1 or 2 includes, according to the feature quantity specified by the image processing parameter selected from the correction instruction, the specified feature amount of the adjustment target parameter specific mechanism is adjusted from the top The parameters of the object. f 13. A method for adjusting a parameter, comprising: a target image creation step of 'receiving an instruction to correct a result image produced as a result of image processing from an operator, and producing the correction instruction a target image reflected after the result image, and a parameter adjustment step of adjusting a parameter value of the image processing to obtain a result image and a target image which are produced as a result of performing the image processing Like the same. 〃 14· A computer-readable recording medium containing a program which can be implemented by a computer to implement the following functions in the computer: an image processing function for performing specified image processing; a target image a production function that receives a correction instruction for a result image produced as a result of the image processing, creates a target image in which the correction instruction is reflected in the result image, and a parameter adjustment function that adjusts the image The parameter values are processed such that the resulting image produced as a result of the image processing described above coincides with the target image. 15. A data processing apparatus, comprising: 124049.doc 200831887 a data processing organization that performs specified data processing; = a data production organization that accepts an instruction to correct the result of the processing of the data, and makes The above amendments refer to: the target data after the above-mentioned result data; and the parameter adjustment mechanism, the parameter value of the bean adjustment and the f:... processing, so that the data obtained by the result of the power treatment and the above target The data processing device of claim 15, wherein the data processing unit performs image processing of comparing the reference image with an inspection target image obtained by outputting the reference image, and checking the inspection object a portion of the defect generated in the image, and image data for displaying the detected defective portion in a display form different from the non-defective portion is prepared and obtained as the result data; the parameter of the data processing specifies the defect portion Detection sensitivity 0 17. The data processing device of claim 15 or 16, wherein the above item The standard data production organization displays the above-mentioned result data on the display screen, and inputs the correction indication of the result data from the above-mentioned display surface. 18. The data processing apparatus of claim 17, wherein the parameter adjustment mechanism causes the upper sputum yan: u &amp; changes from the value of the upper parameter, and judges the result obtained under each parameter value. If the 欠 贝 贝 疋 一致 一致 与 与 与 与 与 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋 疋The data processing device of claim 18, wherein the parameter value of the result data determined by the parameter adjustment means of the obtained set is the median. 20. The data processing device of the item 18 is determined by the above parameter adjustment mechanism to determine the adjustment value of the above parameters as The parameter value when the result data of the above target data is initially obtained in the process of changing the value of the parameter. The data processing device of claim 18, further comprising: a correction range characteristic mechanism that specifies a range in which the value of the parameter is changed based on the input set value. 22. The data processing apparatus of claim 18, further comprising: a specific mechanism in the branching, which specifies a change amount when the parameter value is changed according to the input set value. The data processing device of claim 17, wherein the parameter adjustment mechanism uses a genetic algorithm to obtain a combination of result data that is consistent with the target data from a combination of more than one parameter value, and determines an adjustment value of the parameter. To form the values of the combination of the couch. The data processing device of claim 23, further comprising: a depletion-specific mechanism that specifies a range in which the value of the parameter is changed based on the input set value. 25. The data processing apparatus of claim 23, further comprising: the odorizing ridge mechanism </ RTI> responsive to the input set value, the amount of change in the parameter value of the change 124049.doc 200831887 is specified. (9) The data processing device of claim 15 or 16, which comprises: ::: a parameter-specific mechanism 'in accordance with the designated data from the above-mentioned correction instruction, the county, Dan, and the parameters from the above data processing are Adjust the parameters of the object. γ 疋 疋 27 27 27 27 27 27 27 27 27 27 27 27 27 27 27 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 参数 目标 目标Repair, indicating the target data reflected in the above result data; and &gt; numbering step, adjusting the value of the parameter of the above data processing, so that the target data of the Han Dynasty as the result of the processed domain data is consistent. - a brain-readable recording medium containing a program that can be executed by a computer to implement the following functions in the computer: data processing function, which performs specified data processing; target data creation function, Accepting an instruction to correct the result obtained as a result of the processing of the result of the processing of the result, and generating a target data for reflecting the correction instruction in the result data; and a parameter adjustment function for adjusting the value of the parameter of the data processing to Make the information obtained as a result of the above data processing consistent with the above target data 0 ' 124049.doc