JPH03501785A - Optical evaluation method and device for unknown objects - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed description of the invention] Optical evaluation method and device for unknown objects This application is a conversion of U.S. Patent Application No. 920.513, filed October 17, 1986. This is a continuation in part of an optical processing system application filed on December 23, 1987. Relating to Means and Methods for Controlling Optical Inspection Systems of U.S. Patent Application No. 137,464 do. The disclosure of said application is incorporated herein by reference.

(Technical field of invention) The present invention provides an optical object inspection system, particularly an optical inspection system for detecting unknown objects. The present invention relates to a method and apparatus for evaluating agreement with predetermined characteristic values. That strange thing A characteristic value indicator (characterist) of each object to be examined based on transformation. iclIgniNIrex (hereinafter the same), and change this characteristic value indicator. Compare with predetermined characteristic value indicators. For example, the transformed image is a Fourier transformed image. Therefore, characteristic value labels for known and unknown objects are created using the same equipment and method. Ru.

(Background art of the invention) Mechanical inspection or inspection systems are extremely important components in integrated manufacturing systems. It became a component. These can sort, wrap, and resolve defects without human intervention. can be analyzed. For example, by inspecting holes that have been drilled, Be able to determine if the drill bit is worn out.

Many machine vision systems are serial (hereinafter the same) or - Based on digital electronic technology using dimensional processing.

The image is captured and stored as a matrix of electrical signals. Next, the statue is pre-treated. to enhance edges, improve contrast, and otherwise make the eye visible. Separate objects. The comparison function compares the intensified image with l or more stored images. Compare with the reference image. These preprocessing and comparison functions are typically performed bit by bit ( bit-by-bit) or standard microelectronics based on vector paces. Ronic is done by digital equipment. Therefore, the technology is typically serial Al, which is inherently -dimensional, but the image being processed is two-dimensional. this The dichotomy results in very powerful processing - for dimensional digital devices. It is particularly difficult to do so, and even with enormous amounts of storage and processing power, it cannot be completed completely. It takes a relatively long time to complete.

Digital processing hardware has been enhanced, and software and algorithms have Improvements have been made to the machine viewing system.

However, these improvements reduce system complexity, cost, and programming. Even at the cost of added complexity, you still suffer from the inherent limitations of serial processing. There is.

In some systems, the image to be processed is transformed into a Fourier or other transform. . The Fourier transform provides information about the image of an object by its spatial frequency. A very useful symmetrical pattern to represent. However, the digital combination , -Yu's calculation of the Fourier transform is extremely difficult and requires Micropax (Mic It takes about 1 minute to complete even with a powerful hump coater like ROVAX II. Ru. Power of art array processor Even in a powerful and expensive state, it takes a full second just to create the transformation. modern industry In facilities, production line speeds are often much faster than this.

The computational power and time can be reduced by parallel processing techniques, e.g. the real image of the object being inspected. is converted into a transformed image and optically processed using the available techniques. reduced. Following the generation of the transformed image, the transformed image will occupy that preselected number of spaces. It is "processed" by quantifying the light from a target area, or segment. Next , these quantities are processed electronically to obtain the transformed image and thus the underlying purpose. Provides an abbreviation of an object or a composite property value indicator. Time and cost associated with processing all converted images Compared to It will be evaluated to determine whether it matches the standard.

Inspection systems of the type described above have the advantages described above, but also include other system components, e.g. For example, means for generating electrical signal data representing the appearance of each object to be inspected; means for producing a visible image of each object represented by the received object; from this visible image; In addition to the means for producing a transformed image of the object, means for generating a mark must be included. . One or more samples obtained from a known sample where the target is all good or all bad. is an acceptable or good item by creating more characteristic value indicators. Automatically generate indicators representing predetermined ranges of or discarded items This can be classified by focusing carefully on the nature of the problems that give rise to . This requires, as a prerequisite, a predetermined A simple, fast, and reliable way to provide “signs” or data vectors that represent characteristic values. Requires suitable methods and equipment.

(Summary of the invention) With the above in mind, the present invention provides a method for determining the predetermined properties of an unknown object. Provides an improved method and means for optically evaluating agreement with values. , including the following steps:

That is, the generation of a transformed image of a known object, and the generation of this transformed image from different spatial regions. Sampling light to generate signal data representing different regions, this signal Collect data and collectively define characteristic electrical sample vectors that represent the transformed image of the known object. Memorizing this for the first time, occurrence of optical transformation of unknown object, transformed image of unknown object Signal data representing different regions is generated by sampling light from different spatial regions of the This signal data is collected to generate a characteristic value voltage that represents the transformed image of the unknown object. Collectively define the energy signature vectors and the electrical signals representing unknown and known objects. By comparing the sign vectors, the sign vector of the unknown object becomes the sign vector of the known object. The decision is whether or not it matches the range of .

This method uses light intensity sampling as a way to characterize different regions of the transformed image. sampling and generating optically transformed images of various known objects to determine each region of the transformed optical image. range of values for a region and sampled from different spatial regions. and determining statistical variations between certain amounts of light. This device is a first imaging means for generating electrical signal data representative of the appearance of each object being scanned; a second imaging means for generating a visible image of the object in response to signal data; a third imaging means for generating a transformed image in response to the transformation image; Samples the light of the target object to be inspected and generates electrical signal data representing it. and indicator generating means that collectively define the characteristics Mm. The control means Controls and coordinates the operation of system components to enable extremely rapid and accurate inspection of objects. Ru.

In a preferred embodiment of the apparatus for implementing the present invention, the label of the present testing system The generating means is a movable mask means for obtaining light samples from different regions of the transformed image. including a system control means that controls the actuation of other system components and the movement of the mask means. The computer also synchronizes the intensities of light from different spatial regions of the transformed image. Collect the data that represents the object and assemble this data to create the characteristic value electricity that represents the transformed image of the known object. Determine the sample vector and use the mark vector thus formed to transform the unknown object. Compare with the corresponding indicator vector representing the transformation.

The first imaging means of the inspection system is preferably arranged at a preselected inspection position. Detection means for detecting the presence of an object and illumination of each object at such a position. stroboscopic illumination means and the appearance of objects in such positions. video camera or similar means for generating electrical signals.

The control means initiates the actuation of the stroboscopic illumination means and thereafter illuminates the object in question. was confirmed to be within the field of view of the camera means, and the latter was in the inspection position. from the camera means only after completing the transmission of signal data representing the appearance of the object. Transmission of desired signal data is performed. Furthermore, preferably the control means The operation of the light-sampling and other components of the sign generating means shall be inspected. delay until the optical real and transformed images of the target object reach optimal quality.

Once the optimally transformed image of the object has been generated, optical sampling or masking means are used to detect the transformed image. Separating different spatial regions, the photodetector displays the intensity of light in each different region. generates an electrical signal.

This electrical signal data is stored corresponding to each area, and these correlated data are An electrical signature vector is formed that collectively represents a transformed image of the known object. identical to this The technology is capable of marking known objects and unknown objects, regardless of whether they are "good" or "bad" objects. Used to collect marks of objects. The indicator vector can be used as needed or desired. After further pre-processing, the unknown object is compared and finally the synopsis of the known object is compared. Determine whether the characteristic value matches the determined characteristic value.

Thus, the method simply matches a predetermined mark of an unknown object. It is not only a means of determining, but also a method of organizing and generating signs in the first place. Ru. Markings allow a holistic characterization of the entire appearance of an object.

Once the signs are assembled to form a complete characterization of a good object, acceptance Anything that does not have a suitable label can be discarded. This is during online production time. Allows practice. This method collects many markers and analyzes them as a group. , form a generalization principle of correct signs, and draw a meaningful boundary between good and bad. It is intended to set boundaries.

(Brief explanation of the drawing) Other features of the invention will become apparent from the following description of illustrative embodiments thereof. However, this should be read in conjunction with the accompanying drawings.

FIG. 1 is a schematic diagram of the operating and control components of an optical object inspection system according to the present invention. Figure 2 is an enlarged side elevation view of the rotating mask element of the system, and Figure 3 shows each of the 16 rotating mask elements. a map or graph of the transformed image intensity in the area tr of the object being inspected; im, and Figure 4 shows a number of markers superimposed on each other for multiple target specimens. A representative map or graph showing acceptable ranges for each of the 16 areas. FIG. 1 is a schematic flow sheet illustrating the method of operation of the present invention.

(Description of preferred embodiments) To explain the drawings in more detail, the reference numeral 10 in FIG. 11 indicates a conveyor means. and move the object 12 showing only one object in the direction of arrow 14 to the 'upstream' object inspection position. 16 and a "downstream" target disposal location [18]. guide

Objects 12 are moved onto conveyor 10 at a relatively high speed, typically 15 objects per second. Transported, boxed, canned, bottled, or generally uniform or standardized Processed into other types intended by the type. The object 12 is To ensure that uniformity standards are met, each object reaches the inspection location 16. automatically checked when The object 12 that fails this inspection is placed in the disposal position 18. and is moved laterally from the conveyor IO by the push-out mechanism 20. Pass this test The "good" objects that have been removed are routed by conveyor 10 through location 18 to be dispatched or transported to the desired location. The user is directed to a location (not shown).

The object 12 inspection system generates an electric signal representing the appearance of each object arriving at the inspection location 16. a first imager for generating electrical signal data; and a first imager for receiving the electrical signal data; a second imaging means for generating a visible image of the object; and a second imaging means for generating a transformed image of the object from the visible image. a third imaging means for imaging light from a limited number of different spatial regions of the transformed image; The converted image and thus characteristic value indicators or descriptions of the object to be inspected are a sign generating means for generating electrical signal data defining the main points; and control means for automatically correlating and controlling.

More specifically with respect to said components, the first imaging means may e.g. The arrival of each object 12 at the inspection position 16, such as a mold object detector, is detected and a signal is emitted. a detection means 22 that illuminates the object at the inspection position for a short period of time when activated; A shoop seal lighting means 24 captures a real image of the illuminated object at the inspection position. video camera means 26 for generating electrical signal data representing a real image of the object. .

The second imaging means of the test/stem comprises a frame buffer means 28 and preferably a liquid. spatial light modulation ("SLM") means 30, which is a crystal display device; 71 Norm buffer 28 receives signal data as it is emitted by camera 26. It receives essentially real-time data and retransmits this data to the SLM30 at regular intervals. , causing the SLM to generate a high quality visible image of the inspection object 12. The frame buffer is Typically, it is a fast-access, random-access memory device5, with real image data. Each line of data is memorized as it occurs. All real image data is stored , i.e. when the last horizontal trace is generated and stored, all data are transmitted to the optical modulator. I, I'll do it by the control means of the inspection system. When configured, frame buffer 28 also rapidly transfers the image generated by SLM 30. Extinguish.

The third imaging means of the inspection system is a laser that produces a coherent beam of light. 32 or other means, the light passes through the SLM 30 to form the optical axis of the system. do. As it passes through the SLM, the real image generated by the SLM is reflected in the beam of light. is pressed. The third imaging means further includes lens means 34, which is used for the purpose receiving a modulated light beam carrying a real image of the object 12 and converting the image into a Fourier image of the object to be inspected; Or change to another converted image.

The sampling and indicator generation means of the apparatus are illustratively in the form of a rotating disk 36. It includes movable mask means and light detection and measurement means, such as a light detection device 38. Di The disk 36 is rotated around a central axis at a constant speed by a servo motor 40. This is also shown in Figure 2. Disk 36 has different orientations. A circular arrangement consisting of a wedge-shaped and ring-shaped mask 42 in shape and/or size. It has columns. Each mask represents a different spatial region or segment of the transformed image. However, together they represent a composite or mosaic of the whole picture.

As the disk 36 rotates about its central axis, the mask 42 is continuously exposed to the light detection device. in line with the optical path or line of sight between the vice 38 and the lens means 34. Moving. Thus, the light detection device 38 can be detected via the aperture 42 in the winged disc. The light transmitted to the fourier transform image generated by the lens 34 It is from the target area. Thus, a preselected limited number of transformed images Light from different spatial regions is received by a photodiode device 38; It is converted into an electrical data signal by device 43. Each such signal , from each sampled spatial region of the transformed image by the detection device 38. It represents the intensity of the received light. Collectively these are the transformation images, and thus It represents a characteristic value indicator of the object 12 at the inspection position 16. For example, disk 36 is 1 Contains 32 masks divided into 2 sets of 6 masks. One set of 16 pieces is 16 pieces. Contains a rust or "bowtie" shaped mask; the other set has 16 ring shaped or dona shaped masks. Including a horn-shaped mask. Each wedge-shaped mask has a width of 180716-11.2S arc deformation. and are positioned at different angles in complementary angle segments of 11.25 arc degrees. There is. The 16 wedge masks together form a full image field composite. same Similarly, since each donut shape has a different radius, they also collectively cover the entire image range. form a composite of the surroundings.

The disk also has timing marks 41 on its periphery. this timing The mark may be a mask and may be exposed to light using one of a number of known techniques or detection means 46. to determine the velocity and exact angular position of the disc.

The intensity of the related image area generated by device 38 and defined by each mask 42 The signal data representing the degree is processed by conventional signal processing means 43. this is, Current-voltage converter, sample-hold circuit, analog-to-digital converter, puffer It can include things like means.

The output data from the signal processing means 43 can be received by a computer; Both are part of the control means 44 of the inspection system. Combi Island - The data that Yu receives is , represents the intensity of light for each region of the transformed image. As shown in Figure 3, By graphically illustrating the intensity of the can get. The convenience store stores data, but this system (also The object 12 currently being inspected is judged to be "good" based on its characteristic value indicator data. to determine whether it is within acceptable limits defining a “good” or “bad” category. used for. When it is determined that the 'bad' category is appropriate, the controller 44 A control signal is issued to the extrusion mechanism 20 so that the object 12 in question reaches the disposal position 18. When this happens, it is moved from the conveyor 10 by a push-out mechanism. Also, The controller 44 stores a record of the number of defective objects detected during a particular period of time; An alarm 47 can also be activated when this number reaches a predetermined corner.

In addition to the functions described above, the controller 44 controls the inspection system's three imaging means and marker generation. Automatically controlling and coordinating the operation of each of the various components of the means. Inspection system A typical cycle of operation of the stem begins with inspection position ff1116 and object 12. and the arrival of the camera 26 within the field of view is detected by the detector 22 and reported to the controller 44. be done. Controller 44 may include camera 26, display device or SLM 30.　 The operation of photodetector 38 is continuously monitored.

If the conditions of these components are appropriate for the start of the test system operating cycle, If so, the controller 44 begins its operation upon receiving the signal from the detector 22. do. If one or more of the above components are not in suitable conditions, they The initiation of the test cycle is delayed until suitable conditions are met.

Once the object is in the proper position, the controller 44 is tested for activation of the strobe light 24. Transmission of signal data representing the real image of the object from camera 26 to frame buffer 28 I do. Such data is received and stored by the frame buffer and is Once the real image is stored, it is transmitted to the liquid crystal SLM 30. After that, the SLM of the SLM 30 specified by the controller 44 and used in the inspection system. It is refreshed by the frame buffer for a certain period of time commensurate with the response time. .

onto the light beam created by SLM 30 and then generated by laser 32. The impressed visible image is converted into a transformed image by lens means 34. Di As the mask 36 rotates and a different mask aperture 42 is moved into the optical path of the detector 38, Different spatial regions are detected by photodetector 38.

When the inspection system starts operating, the servo motor 4o moves the disk 36 at a constant speed. Rotation is initiated by controller 44 and then monitored by the controller. De The timing mark 41 on the disc 36 is a tachometer type sensor associated with the disc. It is continuously scanned by means 46 (FIG. 1). The output side of the tachometer 46 is Continuously generates a signal representing the uniform rotational speed of the disk. Also, this signal Instantaneous rotational position of disk aperture 42 relative to the optical path between photodetector 38 and the transformed image confirmation by the controller 44. The signal from the tachometer is transmitted to the photodetector 38 Signal data of the intensity of light emitted by each successive mask opening 4 of the disk 2° (i.e. the transformed image of the spatial domain) by the controller 44. used. Such correlation is determined by the combinator element of controller 44 for each test cycle. Appropriately check the characteristic value indicator signal data generated during the cruise, and then This is necessary for comparison with stored characteristic value indicator signals for "good" objects.

Acceptance of signals from photodetector 38 during each test cycle ．． The image emitted by 34 is delayed until the optimum condition is reached. The period of delay is It depends on the response time of the particular device 30 used in the test system, among other factors. Exists.

One specification of a system where the inspection cycle time for each object was 66 milliseconds. , the transformed image is most stable during the last 24 milliseconds of each cycle. The characteristic value samples of the images were obtained during this final period. Of course, during this period the detector Generation and/or transmission of signal data by 38 is performed by successive mask apertures 42. Allowed by controller 44 only when in line with the optical path between the image and device 38. It is understood that it will be done.

Signs representing the spatial frequency distribution in the transformed image are collected and stored in a sample file. It will be done. The intensity of light for each region of the converted image is determined by the characteristics of the camphor 　res　hereinafter the same). As shown in Figure 3, the intensity points for each area are The connected line represents the marker s itself.

For each sample file, average the corresponding features of all the signs in the sample file. form a synthetic label of ile. The difference between the highest and lowest intensity for each region or the sign A feature is the strength of a domain, or the range within which a feature must fall in order to be accepted. Represents the surrounding area.

In Figure 4, the average label for a group of similar products is indicated by the symbol A. and the range of acceptable intensity for region 1 is indicated by the symbol R. . 41 knowledge 11. A selected subset of data from the sampled transformed image It can be composed of For example, the label ``form'' typically refers to the transformed image. represents the shape of the image generated by the frequency orientation 1 or wedge-shaped mask . The label ``texture crystal'' is determined by the frequency distribution of the transformed image or by the ring mask. It represents the situation that occurred.

Other techniques use 'offsets', which are By having the device deduct the pack ground tan of the included system, be. Background removal is not necessary for effective operation of the system , because if they are generated using the same device, it will be possible to generate both known and unknown objects. This is because both signs include the same background element. these are, The background that exists is identified based on the relative difference between the two. They effectively cancel each other out. To generate an offset specimen, one brush must be Using a blank piece of paper as an object, a marker, typically represented by a straight line, is obtained.

To assess whether an unknown object matches a good set of objects, use a good eye. Pass a group of objects in front of the camera and record their markings to identify good objects. , or the average Hashimoto vector is generated. of a certain situation Basically, it is possible to generate a sample of a known object consisting entirely of defective products. It is also desirable that the nature of the defective product can be identified.

Once a large number of markers for a known object are collected, statistical evaluation becomes possible. the validity of the data collected for one or more areas, or the significance of the variation. It can be evaluated. This is typically a distance metric (distance ml!1rlcl) ! (hereinafter the same)). Another technique is to fit the covariance matrix , which reflects how characteristic values in different regions change in the same direction and amount. The purpose is to collect mathematical data.

As an example of a distance metric, the Euclidean distance to a sign can be calculated using one sample file. Add the square of the difference between each feature value in the sign and the average value of all features of the sign, and find the square root of this sum. It is calculated by Euclidean distance is the separation for all features of a sign It is an accumulation of Although the separation between one feature and the pond feature may not be large, , separation of the total knowledge from one sample file to another sample file can be large and significant. Another useful evaluation is the It is a matter of creating a large separation. This metric is easy to compute and sensitive to cumulative differences between features. It's not a feeling. An average is calculated that represents the average of all labels in the sample file.

The standard deviation is calculated, which is the measure of variation from the mean for each item in the sample. degree. The higher this number, the better the object, whether it is a good object or a waste object. There are large differences between each item in the sample file. Other measures of variation are the eyes of the maximum difference It is a measurement of a target. This is a measure of how far the best waste object in the sample is from the average. reflect how far apart they are. The higher the number, the greater the flatness of all indicators in the file. The difference between the average and the most different sign from the average is large. Other techniques include Chebi Cech metric, Fukanagar-Koontz algorithm, Forey-Solmon algorithm algorithm, Fisher's linear discriminant formula % formula % To make accept/discard decisions, electric transport vectors representing unknown and known objects are used to make accept/discard decisions. The boundary between the sign representing a good object and the sign representing a waste object (t h It is compared with r　a　s　h　o　1 (+, the same applies hereafter).

If the sign falls outside the range defined for different areas of the good object, The specimen range for good samples and the label range for discarded samples. A wider range can be defined by setting boundaries in the middle. child This would set wider standards for acceptance of good objects, but not for disposal. Maintain a minimum distance from signs characterizing the object.

Boundaries can be defined automatically or manually, as necessary or desired. Ru. Through these methods, clusters of signs can be detected and separated at different points in the measurement. Place a boundary between the segments of the metric represented by the group of signs to separate them be able to. The larger the distance or window between the sign representing the good object and the waste object, the better. The probability of accurate discrimination between good objects and waste objects during product inspection is high.

In FIG. 5, the sequence of steps 51-54 includes the optical transformation of a known object. image generation, sampling of the light intensity in the transformation region for each known object; Represents a memory of the intensity for each region of each sampled object. This loop is Continued for the entire group of known objects. for each region of each known object. After storing electrical signal data representing intensity, it is collected to create a transformed image of a known object. A characteristic value electrical sign vector representing . This indicator vector is typically A compilation of all the intensities in the region, where the intensities are An average 1 m vector representing acceptable characteristic values of the Fourier transform for different regions of the object. It can be used to define a range of values.

This is represented by box 55 in the yJ5 diagram.

Steps 56 and 57 include generation of an optically transformed image of the unknown object and transformation of the unknown object. represents the sampling of light from different spatial regions of the image, relative to a known object. It is done in the same way. An electrical signal representing the light sampled from each of its different regions. The signal data is collected to define a characteristic electrical sign vector representing the transformed image of the unknown object. Melt. The electric sign vector representing the transformed image of the unknown object represents the transformed image of the known object. This is compared with the electric sign vector for each region, which is a binary network. can be implemented using standard computer processing techniques. Depending on the desired end result , this comparison shows that the degree of match for the average label, the conversion of the unknown object to the known object The number of regions where the average Hashimoto coincides with the unknown object's sign is the transformed image of the known object. It can express the number of areas that fall within the acceptance range for the area. good object vs. A statistical evaluation of the number of waste objects can also be carried out with the highest degree of variability present. Ru. Inspection is performed by the degree of correlation between the sign vectors for known and unknown objects. The decision to accept or discard an object is expressed by a formula, and an electrical signal indicating it is transmitted to a downstream location to collect or divert unknown objects.

Although the preferred embodiment of the invention has been particularly shown and described, this is for illustrative purposes only. The scope of the invention, and not for purposes of limitation, is defined by the following claims.

Copy of amendment 'D translation) Submitted Article 184-8 of the Patent Act) June 1990, 22 Day Yoshi, Commissioner of the Patent Office 1) Takeshi Moon disaster 1. Display of patent application PCT/US8B104636 2. Name of the invention 3. Patent applicant Address: 1989B Wilmington Market, Praue State, United States Street Street 1007 Names E, I, DuPont Denimoas Country Nationality: United States 4. Agent Address: 4-8 Meieki, Nakamura-ku, Nagoya @ 12 Ryoshin Building 5. Date of submission of written amendment January 26, 1990 6. List of attached documents (1) Copy and translation of written amendment) Scope of request 1. Optically extracts a certain characteristic value of the transformed image of the unknown object and extracts this extracted characteristic value. The unknown object is determined by comparing the characteristic values of the transformed image of the known object A method for optically evaluating the conformity of an object with a predetermined standard, the method comprising (1) ) generation of an optically transformed image of the known object; and (2) generation of the optically transformed image of the known object. By sampling light from different circumferential and radial components of the intermediate region, Regardless of any electrical signals generated in response to the detection of the real image of the object.

(3) generating electrical signal data representing the solid area; Electrical signal data representing different circumferential and radial components of the spatial region of the transformed image of the object. Collectively define characteristic value electrical sign vectors that collectively represent the transformed image of the same known object. remembering that and (4) Generation of an optically transformed image of the unknown object; and (5) Transformed image of the unknown object. By sampling light from different circumferential and radial components of a spatial region of Regardless of any electrical signal generated in response to the detection of a real image of the intellectual object, said device (6) generating electrical signal data representing the area where the unknown target object has become; Collect electrical signal data representing different circumferential and radial components of the spatial region of the transformed image. Collectively determining a characteristic value electric sign vector representing a transformed image of the unknown object; (7) Compare the electric sign vectors representing the transformed images of the unknown object and the known object. In comparison, the label vector for the unknown object matches the label vector for the known determine whether the unknown object qualifies as an acceptable object. determining whether the An optical evaluation method for unknown objects consisting of each step.

2. The step of generating a transformed optical image converts the reflected light from the image of the object into 2. A method according to claim 1, comprising the generation of an optically transformed image.

3. The above-mentioned step that samples the light and generates electrical signal data representing a transformed image of the target object. samples light from many circumferential and radial components of the spatial region of the transformed image. and emit an electrical signal representing the intensity of light in each sampled area. 2. The method of claim 1, which comprises:

4. Collect the electrical signal data for the known object and determine a marker vector. the step of storing the electrical signal data in the spatial region it represents; Correlate the circumferential and radial components of Claim 1, comprising: storing in a predetermined order according to an area; How to put it on.

5. Light from different circumferential and radial components of the spatial region of the transformed image of the unknown object. The step of sampling from the transformed image of the known object The space of the transformed image of the unknown object corresponding to the circumferential and radial components of the spatial region claim consisting of sampling light from different circumferential and radial components of a target area The method described in Section 1.

6. Before collecting the electrical signal data for the unknown object and determining the marker vector. the electrical signal in the same manner as the step was performed for the known object. 2. The method of claim 1, comprising collecting data and defining an indicator vector.

7. Collect the electrical signal data for the known object and determine a marker vector. the step of storing the information obtained from the collected electrical signal data; Define an In vector containing any of the previously determined range of acceptable data. 2. A method according to claim 1, comprising storing the same.

8. Generate electrical data signals representing differences between sign vectors of many known objects. 2. The method of claim 1, further comprising the step.

9. said step of generating an optically transformed image of a known object comprises said step of collecting light and generating signal data, including generating a chemical transformation, said plurality of steps collecting light to a known object of the invention and generating signal data; Said step of gathering data and defining and storing indicator vectors includes said step of 4[ 2. A method as claimed in claim 1, comprising defining and storing a vector.

10. The step of collectively defining the indicator vectors Generate and determine statistical data representing predetermined variations in data. 10. The method according to claim 9, further comprising obtaining an optimal value based on the variation determined. the method of.

11. The step of generating statistical data is based on differences in the transformed image of the object. From determining the Euclidean distance for the signal data in the spatial domain The method according to claim 10.

12. The step of generating statistical data is based on differences in the transformed image of the object. 11. The method of claim 10, further comprising determining a maximum separation between features in a spatial region of Method described.

13. Determine a marker vector representing the composite transformed image for the known object and use it as Said step of storing accepts the circumferential and radial components of the spatial region of the transformed image. Define and store a single indicator vector that represents a range of signal data values, and Since the range is defined between the highest and lowest signal data for said spatial region. The method according to claim 9.

14. Optically extract a certain characteristic value of the transformed image of the unknown object and use this extracted characteristic The unknown object is determined by comparing the values with the corresponding characteristic values of the transformed image of the known object. means for optically evaluating the conformity of an object with a predetermined standard, 1) means for generating an optically transformed image of the object; and (2) a space for the transformed image of the object. By sampling light from different circumferential and radial components of the target area, Regardless of any electrical signals generated in response to the detection of a real image of the object, the (3) means for generating electrical signal data representing a region; and (3) a spatial region of the object. Collect electrical signal data representing different circumferential and radial components to create a transformed image of the same object. (4) means for collectively determining and storing characteristic value electrical sign vectors to represent; Compare the electric sign vectors representing the transformed images of the unknown object and the known object to determine the same unknown object. Whether the target's sign vector matches the 1m vector for the same known target. determine whether the unknown object qualifies as an acceptable object. means to determine whether Optical evaluation means for unknown objects consisting of

15. The means for generating a transformed optical image uses light reflected from the image of the object. 15. Apparatus according to claim 14, comprising means for generating an optically transformed image.

16. Before sampling the light and generating electrical signal data representing the transformed image of the target object A recording means samples light from a number of angular and radial components of the spatial region of this transformed image. emit an electrical signal representing the intensity of light in each different area sampled. 15. The apparatus according to claim 14, comprising means for generating.

17. The method of collecting the electrical signal data, determining a marker vector, and storing it. A stage converts the electrical signal data into angular and radial components of the spatial region it represents. Correlate and organize this correlated signal data according to the region it represents. 15. The apparatus of claim 14, comprising storing in a determined order.

18. From different angular components and radial components of the spatial region of the transformed image of the unknown object. The means for sampling light from a transformed image of the known object The space of the transformed image of the unknown object corresponding to the angular and radial components of the spatial region claim consisting of means for sampling light from different angular and radial components of a target area; The device according to item 14.

19. Said method of collecting said electric signal data, determining a sign vector, and storing it. a stage with predetermined acceptance obtained from said collected electrical signal data. Is it possible to determine a sample vector that includes any of the data ranges that can be used and store it? 15. The apparatus of claim 14, comprising:

20. Generates electrical data signals representing differences between sign vectors of many known objects 15. The apparatus of claim 14, further comprising means for.

21. The means for generating an optically transformed image of a plurality of objects image generating means, said means for collecting light and generating signal data being used for a number of purposes; a means for collecting light and generating signal data; and a means for collecting the signal data; Said means for defining and storing sign vectors collects signal data from said plurality of objects. Collect data to determine a sign vector representing a composite transformation image for the same object and store it. 15. A device according to claim 14, comprising means for:

22.VI recording means for collectively defining the indicator vector, collected electrical signal data Generates statistical data representing variation between predetermined and determined 22. The method of claim 21, further comprising means for obtaining an optimal value based on the variation. Device.

23. Determine a marker vector representing the composite transformed image for the known object and use it as said means for storing are acceptable for the angular and radial components of the spatial domain of the transformed image; A signal? - define and store a single indicator vector representing a range of data values; and the range is defined between the highest and lowest signal data for said spatial region; 22. The apparatus of claim 21.

international search report

Claims (1)

[Claims] 1. A certain characteristic value of the transformed image of the unknown object is optically extracted, and this extracted characteristic value is The unknown object is determined by comparing it with the corresponding characteristic value of the transformed image of the known object. A method for optically evaluating agreement with a predetermined standard, the method comprising: (1) (2) generating an optically transformed image of the known object; and (2) generating a different optically transformed image of the known object. Electrical signal data representing the different regions by sampling light from spatial regions to occur, and (3) Collect electrical signal data representing different regions of the known object to obtain the same known object. To collectively determine and store characteristic value electric sign vectors representing the transformed image of objects. (4) generating an optically transformed image of the unknown object; and (5) generating a transformed image of the unknown object. by sampling light from different spatial regions of the electrical field to represent said different regions. generating signal data; (6) Collect electrical signal data representing different regions of the unknown object to obtain the same object. collectively determining a characteristic value electric sign vector representing a transformed image of the object; (7) Compare the electric sign vectors representing the transformed images of the unknown object and the known object. and the tag vector for the unknown object matches the tag vector for the known object. determine whether the unknown object qualifies as an acceptable object. to determine whether An optical evaluation method for unknown objects consisting of each step. 2. said step of generating a transformed optical image comprises an optical transformation from a visible image of the object. 2. The method of claim 1, comprising generating an image. 3. The screen samples the light and generates electrical signal data representing a converted image of the object. Theabu samples light and analyzes the light in each sampled area. 2. The method of claim 1, comprising generating an electrical signal representative of the intensity. 4. Collecting the electrical signal data for the known object and determining a marker vector. The step of storing the electrical signal data correlates the electrical signal data with the region it represents. and predetermined this correlated signal data according to the region it represents. 2. A method according to claim 1, comprising storing in a defined order. 5. sampling light from different spatial regions of the transformed image of the unknown object; The step includes converting the transformed image of the known object into a spatial region that has been sampled. Sampling light from different spatial regions of the corresponding transformed image of the unknown object. 2. The method according to claim 1, comprising: 6. Before collecting the electrical signal data and determining the marker vector for the unknown object. the electrical signal in the same manner as the step described above was performed for the known object 2. The method of claim 1, comprising collecting data and defining an indicator vector. 7. Collecting the electrical signal data for the known object and determining a marker vector. the step of storing the information obtained from the collected electrical signal data; Define and store an indicator vector containing a previously determined range of acceptable data. 2. The method of claim 1, comprising: 8. Generates electrical data signals representing differences between sign vectors for many types of known objects. 2. The method of claim 1, further comprising the step of generating. 9. The step of generating an optically transformed image of a known object comprises said step of collecting light and generating signal data, including generating a chemical transformation, said plurality of steps collecting light to a known object of the invention and generating signal data; Said step of gathering data and defining and storing indicator vectors includes said step of A sign that collects signal data from a known object and represents a synthetic transformation image for the same known object. 2. A method as claimed in claim 1, comprising defining and storing a vector. 10. The step of collectively defining the indicator vectors is performed using the collected electrical signal data. Generates statistical data that represents variation between predetermined 10. The method of claim 9, further comprising obtaining an optimal value based on the variation determined. Method. 11. The step of generating statistical data may include different transformations of the transformed image of the object. from determining the Euclidean distance for signal data in the spatial domain. 11. The method according to claim 10. 12. The step of generating statistical data may include different transformations of the transformed image of the object. 11. The method according to claim 10, comprising determining a maximum separation between features in a spatial domain. How to put it on. 13. Define and record a marker vector representing a composite transformed image for the known object. the step of storing acceptable signal data values for each spatial region of the transformed image; define and memorize a single marker bacter representing the range of the space, and the range is Claim 9: defined between the highest and lowest signal data for the target area. The method described in. 14. Optically extract a certain characteristic value of the transformed image of the unknown object and extract this extracted characteristic value. The unknown object is determined by comparing the characteristic values of the transformed image of the known object A means for optically evaluating the conformity of an object with a predetermined standard, the method comprising (1) (2) means for generating an optically transformed image of the object; and (2) means for generating an optically transformed image of the object. Electrical signal data representing the different regions is obtained by sampling the optical hairs from the intermediate regions. the means by which it occurs; (3) Converting the same object by collecting electrical signal data representing different areas of the object means for collectively determining and storing characteristic value electrical sign vectors representing the image; (4) Comparing the electric sign vectors representing the transformed images of the unknown object and the known object. The mark vector for the same unknown object matches the boundary of the mark vector for the same known object. determine whether the unknown object is an acceptable object. Means to determine eligibility Optical evaluation means for unknown objects consisting of 15. The means for generating a transformed optical image is configured to generate an optical transformed image from a visible image of an object. 15. The apparatus of claim 14, comprising a biochemical means. 16. The above-mentioned method samples the light and generates electrical signal data representing a transformed image of the target object. The means samples the light and determines the intensity of the light in each different area sampled. 15. The apparatus of claim 14, comprising means for generating an electrical signal indicative of the temperature. 17. said means for collecting said electrical signal data and determining and storing a marker vector; correlates the same electrical signal data with the region it represents, and this correlated signal storing data in a predetermined order according to the area it represents 15. The apparatus of claim 14, comprising: 18. sampling light from different spatial regions of the transformed image of the unknown object; the means for determining the spatial region sampled from the transformed image of the known object; sampling light from different spatial regions of the corresponding transformed image of the unknown object; 15. The apparatus of claim 14, comprising means. 19. said means for collecting said electrical signal data and determining and storing a marker vector; is a predetermined acceptable value obtained from the collected electrical signal data. A claim consisting of determining and storing an indicator vector containing a range of data 15. The device according to 14. 20. electrical data signals that represent the differences between sign vectors for many types of known objects. 15. The apparatus of claim 14, further comprising means for generating. 21. said means for generating an optically transformed image of a plurality of objects; generating means for collecting light and generating signal data, said means for collecting light and generating signal data for a number of purposes. means for collecting light and generating signal data; Preliminary means for determining and tracking the signal vector Collect data to determine a marker vector representing a composite transformation image for the same object and store it. 15. A device according to claim 14, comprising means for: 22. Said means for collectively defining a signature vector comprises: Generate statistical data that represents variation between predetermined 22. The apparatus of claim 21, further comprising means for obtaining an optimal value based on the Place. 23. Define and record a marker vector representing a composite transformed image for the known object. said means for storing a range of acceptable signal data values for each spatial region of the transformed image; Define and store a single marker vector representing the spatial region, and the same region Claim 21: defined between the highest and lowest signal data for the range. The device described in.