JP3134776U - Luminescence analyzer - Google Patents

Abstract

[PROBLEMS] To improve the accuracy and efficiency of analysis by always properly detecting a defective portion of a sample surface even if there is a difference or variation in the component type of the sample to be analyzed, the surface pretreatment method, or the imaging system of the sample surface. .
Image data of a sample surface imaged by a CCD camera is transmitted to a personal computer for image analysis. The personal computer 3 obtains the average luminance of the area excluding the portion where the luminance is lower than the provisional threshold for each judgment group name stored for this image, and calculates the relative threshold (the difference threshold or the ratio threshold based on the ratio) from the average luminance. A portion whose luminance is smaller than the threshold value specified in the image is detected as a defective portion. The analysis position is extracted based on the defect information, and the robot arm 4 is controlled and moved based on the table of the extracted analysis position, the sample is positioned, and the analysis position is analyzed.
[Selection] Figure 1

Description

  The present invention relates to various analyzers that analyze the surface of a solid sample, such as a light emission analyzer and a laser ICP (inductively coupled plasma) analyzer that detect and analyze light from a sample by emitting light by discharge.

  An emission analysis apparatus as an example of an apparatus for analyzing the surface of a solid sample performs analysis by switching the type of discharge such as spark discharge or arc discharge depending on the shape or material of the solid sample (hereinafter referred to as sample). When analyzing a solid sample, the sample is placed in the analysis gap and discharged by flowing a high current. However, if a discharge is performed on a defective portion of the analysis surface (hereinafter referred to as the sample surface), the emission level is different from the normal portion, An accurate analysis is not possible.

In order to perform an accurate analysis, there is provided an emission analyzer that detects a defect portion on a sample surface and automatically avoids the defect portion. In other words, an image of the sample surface is captured and loaded into a personal computer, the image analysis is used to detect the defective part from the difference in brightness from the normal part of the sample surface, and the analysis position is extracted so that the analysis position avoids the defective part. To do. Based on the extracted coordinates of the analysis position, the robot arm is controlled by a personal computer to position the sample and analyze the analysis position (see, for example, Patent Document 1).
JP 2005-69853 A

  When a defective part is detected from a difference in luminance from a normal part of the sample surface by image analysis of the captured image of the sample surface, the determination criterion is a defect determination luminance threshold registered in advance. A portion of the image whose luminance is smaller than this threshold is determined as a defective portion and detected. Conventionally, one type of defect determination brightness threshold has been registered, but the component type of the sample to be analyzed, the surface pretreatment method, the number of treatments after cutting abrasive replacement for surface pretreatment, the sample surface inclination, Since the brightness of the sample surface differs depending on the difference or variation in the brightness of the lamp light source or the sensitivity of the camera, there is a high possibility of erroneous determination with one defect determination brightness threshold (determination criterion).

  Therefore, the present invention is designed to analyze the component type of the sample to be analyzed, the surface pretreatment method, the number of treatments after replacement of the cutting abrasive for surface pretreatment, the inclination of the sample surface, the brightness of the lamp light source, or the camera sensitivity The purpose is to always detect defects on the sample surface appropriately even if there are differences or fluctuations, and to improve analysis accuracy and efficiency.

  Image analysis means for imaging a sample surface of a sample to detect a defective part, means for extracting an analysis position based on information on the defect part, and an analysis part for analyzing the sample at the extracted analysis position In the emission analyzer, means for detecting an average luminance of a region of the sample surface excluding a portion where the luminance is smaller than a provisional threshold and setting a determination criterion relative to the average luminance, and based on the determination criterion And detecting means for detecting the defective portion. Therefore, it is possible to detect a defective portion based on a determination criterion corresponding to the difference / variation between the sample and the imaging system.

  According to the present invention, a relative threshold value from the average luminance of the sample surface is set for each sample type as the defect determination luminance threshold value (determination criterion). Therefore, differences and fluctuations in the component type of the sample to be analyzed, the surface pretreatment method, the number of treatments after cutting abrasive replacement for surface pretreatment, the inclination of the sample surface, the brightness of the lamp light source, or the sensitivity of the camera Therefore, even if the brightness of the sample surface is different, if the relative threshold (judgment criterion) from this average brightness is used, the influence of the difference / variation is corrected and the defective part is properly detected, and the normal part is always detected. The analysis accuracy can be improved and the analysis efficiency can be improved.

  An image of the sample surface is taken by irradiating the sample surface with light from a lamp that emits visible light. The sample surface image is captured by a CCD camera, a CMOS camera, or the like.

  Embodiments of the present invention will be described below with reference to FIGS. FIG. 1 is a configuration diagram schematically showing an embodiment. FIG. 2 is an example of an information table in which a relative threshold value from the average luminance is stored. FIG. 3 is a diagram for explaining the extraction of the analysis position on the sample surface according to the embodiment. FIG. 4 is a block diagram showing an embodiment. FIG. 5 is a flowchart showing the operation of the embodiment.

  As shown in FIG. 1, the emission analysis apparatus of the present invention has an imaging unit having an analysis unit 1 having an electrode for discharge, a CCD camera 21 provided near the analysis unit 1, a lamp 22, and a shielding cover 23. Unit 2, a robot arm 4 as a transport mechanism for holding and transporting a sample, a CCD camera 21, a robot arm 4, and an analysis unit 1 connected to perform analysis processing of image data from the CCD camera 21, A personal computer 3 is provided as a control device for controlling the operation of the robot arm 4 and the analysis unit 1 based on the result.

In the imaging unit 2, a CCD camera 21 is installed upward in order to capture an image of the sample surface, and a lamp 22 is installed in the vicinity of the CCD camera 21 so as to irradiate the entire surface so that no shadow is formed on the sample surface. . As long as the lamp 22 can uniformly irradiate the surface of the sample, it is of course possible to use a donut type or a coaxial incident type coaxial with the camera.
In addition, the imaging unit 2 is surrounded by a shielding cover 23 that prevents the incidence of light that interferes with imaging.

  Image data captured by the CCD camera 21 is transmitted to the personal computer 3. The personal computer 3 performs image analysis on the image data of the sample surface captured from the imaging unit 2 to detect a defect portion on the entire sample surface, extracts an analysis position based on the defect portion information, and extracts the analysis position of the extracted analysis position. Based on the sitting surface, the robot arm 4 is controlled and moved to position the sample.

  For detection of a defective portion on the entire sample surface, a temporary threshold set for each determination group name stored in the information table 31 shown in FIG. 2 and a relative threshold from the average luminance are used. There are two relative threshold values (judgment criteria) based on the average luminance, that is, a threshold value based on a difference and a threshold value based on a ratio. The operator designates at which threshold a defective portion is detected. The threshold based on the difference is determined by a subtraction formula in the corresponding cell when the average luminance is obtained, and similarly, the threshold based on the ratio is determined by a multiplication formula in the corresponding cell when the average luminance is obtained. The information table 31 is stored in the personal computer 3, and the determination group name is determined by the sample component type, the surface pretreatment method, the sample diameter, the required number of analysis points, and the like, and is designated by the operator.

  In detection of a defective portion on the entire sample surface, first, the personal computer 3 obtains an average luminance of a region excluding a portion of the image of the sample surface subjected to image analysis whose luminance is smaller than the provisional threshold. Next, a portion in which the luminance of the sample surface image captured and analyzed by the personal computer 3 is smaller than the relative threshold value from the average luminance (threshold value by difference or specified threshold value by ratio) is detected as a defective portion. The In the image analysis, the intensity of the brightness of each part in the image is digitized in, for example, 255 levels. The intensity of the luminance to be digitized is calibrated by taking an image of a luminance standard sample, taking it into the personal computer 3, and analyzing the image.

  The personal computer 3 controls the analysis unit 1 and the robot arm 4 so that the sample held by the robot arm 4 is analyzed at the analysis position extracted based on the defect part information. At this time, the coordinates of the analysis position are associated with the position coordinates of the robot arm 4 so that the extracted analysis position can be analyzed accurately.

  Image analysis is performed as shown in FIG. (A) is an image of the sample surface taken from the imaging unit 2, and defective portions of the defect areas 41 and 42 are detected in the photographed sample 40. A plurality of analysis candidate ranges 43 are set on the sample surface of the photographed sample 40 as shown in (B), and it is determined by image analysis whether each analysis candidate range 43 is suitable for analysis. In this determination, the suitability of the analysis candidate range 43 is determined by comparing the ratio (%) of the defect portion of the defect areas 41 and 42 to the analysis candidate range 43 with a set value. A range that does not overlap each other is extracted from the analysis candidate range 43 in which the proportion of the defective portions 41 and 42 is equal to or less than the set value, and is recognized as the analysis target position 43a as shown in FIG. Based on the result, the robot arm 4 is controlled.

  The setting of the analysis candidate range 43 is determined by determining the optimal position (radius r) for analysis from the radius of the photographed sample 40 based on the result of image analysis. In addition, the number of analysis target positions 43a, a setting value (%) for determining the suitability of the analysis candidate range 43, and a numerical value for obtaining a relative threshold value from the provisional threshold value and the average luminance for each determination group name (in the case of a difference threshold value, average luminance A parameter such as a numerical value to be subtracted from the numerical value multiplied by the average luminance in the case of a threshold value based on a ratio can be arbitrarily set by the operator.

  The embodiment will be described with reference to the block diagram of FIG. The operator holds the sample on the robot arm 4 and sets various analysis conditions such as the determination group name of the sample to be analyzed and the specification of whether to use a threshold value based on a difference or a threshold value based on a ratio as a relative threshold value from the average luminance. Set and input analysis start to the personal computer 3. The personal computer 3 controls the robot arm 4 based on an instruction from the operator and transports the sample to the imaging unit 2. In the imaging unit 2, the sample surface is irradiated with light, and the image is captured by the CCD camera 21. The captured image of the sample surface is taken as data into the personal computer 3 and subjected to image analysis, and the average luminance of the area excluding the portion where the luminance of the image is smaller than the provisional threshold is obtained. Next, a portion where the luminance of the image is smaller than the relative threshold value from the average luminance (threshold value determined by difference or threshold value specified by ratio) is detected as a defective portion. At this time, the provisional threshold value corresponding to the determination group name of the sample under analysis and the relative threshold value from the average luminance are selected and used. The personal computer 3 extracts the analysis position based on the defect part information obtained by the image analysis, moves the robot arm 4 to the analysis part 1 so as to analyze the analysis position, and performs analysis at the analysis position. Control and capture the analysis data.

  The operation of the embodiment will be described with reference to FIG. The robot arm 4 holding the sample moves the sample to the imaging unit 2, and the imaging unit 2 images the sample surface (step S1). The captured image data is taken into the personal computer 3 and subjected to image analysis, and the average luminance of the area excluding the portion where the luminance of the image is smaller than the provisional threshold is obtained. Next, a portion where the luminance of the image is smaller than the relative threshold from the average luminance (threshold specified by difference or threshold specified by ratio) is detected as a defective portion of the defect areas 41 and 42 (step S2). By detecting the defective portions of the defect areas 41 and 42, several analysis target positions 43a are extracted (step S3). The personal computer 3 controls the robot arm 4 based on the coordinates of the analysis target position 43a, and positions the sample so as to analyze the analysis target position 43a (step S4). When the positioning of the sample is completed, the analysis electrode of the analysis unit 1 is discharged and analysis is performed (step 5). When the analysis is completed at one analysis target position 43a, the robot arm 4 is moved so as to analyze the next analysis target position 43a, and the sample is positioned and analyzed. In this way, the positioning and analysis operations are repeated until all the analysis target positions 43a are analyzed, and when the analysis of all the analysis target positions 43a is completed, the analysis of the sample is ended (step 6).

  Since the present invention is configured as described above, when analyzing different samples such as the component type and the surface pretreatment method, the number of treatments after cutting abrasive replacement for surface pretreatment, the inclination of the sample surface, the lamp light source Even if the brightness of each sample surface differs due to differences in brightness, camera sensitivity, etc., the surface of the sample surface excluding the part with a brightness smaller than the provisional threshold set for each judgment group name corresponding to each sample. The average brightness is obtained, and the relative threshold from the average brightness set for each judgment group name (the threshold specified by the difference or the threshold specified by the ratio) is used to properly detect the defective part. Analysis is possible, and improvement in analysis accuracy and efficiency can be expected.

  In the embodiment, the image analysis shown in FIG. 3 automatically proceeds and the plurality of analysis target positions 43a are automatically analyzed. The image shown in FIG. 3 is displayed on the monitor of the personal computer 3, and this is displayed. The present invention can be applied only when the operator can observe and start the analysis, and only when the personal computer 3 is instructed to start the analysis, the present invention can be applied. Examples are included.

  The present invention relates to various analyzers for analyzing the surface of a sample, such as an emission analyzer and a laser ICP (inductively coupled plasma) analyzer that detect and analyze light emitted from a sample by discharge.

It is a block diagram which shows an Example schematically. It is an example of the information table in which the relative threshold from the average luminance is stored. It is a figure for demonstrating extraction of the analysis position in the sample surface of an Example. It is a block diagram which shows an Example. It is a flowchart figure which shows operation | movement of an Example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Analysis part 2 Imaging part 3 Personal computer 4 Robot arm 21 CCD camera 22 Lamp 23 Shielding cover 31 Information table 40 Photographed sample 41 Defect area 42 Defect area 43 Analysis candidate range 43a Analysis object position

Claims (1)

  Image analysis means for imaging the analysis surface of a solid sample to detect a defective part, means for extracting an analysis position based on information on the defect part, and an analysis part for analyzing the solid sample at the extracted analysis position A means for detecting an average luminance of a region excluding a portion of the analysis surface of the solid sample whose luminance is smaller than a provisional threshold and setting a determination criterion relative to the average luminance; An emission analyzer characterized by comprising a detecting means for detecting the defective portion based on a reference.

Images