JP2008241424A - Image processing apparatus and method using the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable users inexperience in image inspection to easily and speedily configure inspection settings in an image processing apparatus for processing images acquired by imaging. <P>SOLUTION: The image processing apparatus is provided with a storage part in which a plurality of inspection programs are stored; an inspection setting part to be operated by a user to configure desired inspection settings; and a control part for controlling the overall apparatus. The inspection setting part is provided with a data table previously, including a plurality of setting items. The control part automatically sets inspection programs suitable for inspection standards of final setting items of the user (S214 and S216), when the user determines each setting item by a first determination step (S211) for selecting items of inspection objectives; a second determination step (S212) for selecting characteristics of images of objects; and a third determination step (S 213 and S215) for selecting items of inspection standards. Since there is thereby no need for the user to directly select inspection programs, it is possible for inexperienced users to easily and speedily configure inspection settings. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image processing apparatus capable of performing image processing on image data of an object captured by a camera and performing product inspection such as appearance inspection of the object, and an image processing method using the image processing apparatus.

  Conventionally, this type of image inspection apparatus mainly captures an object with a camera in order to identify the type of parts, detection of mixing of different parts, automation of assembly work and product inspection in the product manufacturing process, By measuring and recognizing the characteristics of image data such as the area, shape, and position of an object, it is used to determine non-defective products and defective products. In such an inspection, the inspection targets of image processing are various, and the inspection purposes are diverse. In order to realize these inspection purposes, many processing algorithms are necessary, and there are optimum processing algorithms for each inspection target. Therefore, in such an image processing apparatus, the user must set an optimal processing algorithm that suits the desired inspection purpose before starting the inspection. For this reason, the user must first select a processing algorithm suitable for the inspection purpose by a predetermined inspection setting operation.

  By the way, as a method for selecting a processing algorithm in a conventional image inspection apparatus, for example, a method for selecting each processing algorithm by assigning a name (for example, edge inspection, matching inspection, etc.) that represents the characteristics of the image processing to each processing algorithm, There has been a method of selecting a suitable application by name (for example, an average gray value). However, these methods allow users who are unfamiliar with image processing to inspect the inspection target using any processing algorithm, even if they understand the inspection target and purpose and desired inspection result. The selection of whether a necessary inspection purpose is achieved depends on experience and know-how, and therefore selection of a processing algorithm has been a difficult task. Other methods include lists of inspection targets and inspection objectives frequently used in certain industries (for example, inspection lists for each electrical component in the electrical industry, inspection lists for each container of food used in the food industry, etc.) There is a method to make it correspond to a processing algorithm by making it choose. However, this method has become a user-friendly selection method, but it is difficult to cover all the vast inspection objects and inspection purposes in each industry, and it is necessary to narrow down the inspection objects and inspection purposes. In some cases, the selection range of the inspection was limited and did not apply to the inspection purpose.

Further, as a conventional image inspection apparatus, in an image processing apparatus that visually inspects an inspection target article by collating an input image corresponding to the inspection target article with a reference image according to a collation program set by a user, It is known that necessary operation guidance is automatically performed by inputting a type, and an image collation program suitable for inspection is set by this operation guidance (see, for example, Patent Document 1). However, in this device, a beginner first selects an inspection type, understands the content of the inspection corresponding to the inspection type in the description such as illustrations, and then selects an inspection method corresponding to the selected inspection type. In the same manner as described above, an image collation program necessary for inspection is set by selecting after understanding with illustrations. For this reason, there are many selection steps until the user sets the image collation program, and there is a problem that it takes time to determine the inspection setting.
JP 2001-165863 A

  The present invention has been made to solve the above problems, and an image processing apparatus capable of easily and quickly setting an inspection program in an inspection setting even for a user unfamiliar with image inspection, and an image using the same An object is to provide a processing method.

  In order to achieve the above object, an invention according to claim 1 is an image processing apparatus for inspecting an object based on an image of the object imaged by an imaging unit, wherein the storage unit stores a plurality of inspection programs. A test setting unit for performing desired test settings by a user operation, and a control unit for controlling the entire apparatus. The test setting unit is provided with a data table including a plurality of setting items in advance. When inspection setting of an object is performed, after performing a determination step of selecting one setting item from a plurality of setting items of the data table a plurality of times, the control unit is selected at the end of the determination step. The inspection program corresponding to the set item is called from the storage unit and the inspection is executed.

  According to a second aspect of the present invention, in the image processing apparatus according to the first aspect, the setting items include an item indicating an inspection purpose, an item indicating an image characteristic of the target object, and an item indicating an inspection standard of the target object. The determination step includes: a first determination step for selecting an item indicating the inspection purpose; a second determination step for selecting an item indicating a feature of the image of the object; and And a third determination step for selecting an item indicating the inspection standard.

  The invention according to claim 3 is an image processing method for inspecting an object based on the image of the object imaged by the imaging unit, and a plurality of items indicating an inspection purpose when performing inspection setting of the object A determination step for selecting one item from the above, a determination step for selecting one item from a plurality of items indicating the characteristics of the image of the object, and an item indicating the inspection criteria of the object, by being selected And a determination step of selecting one item from a plurality of items for which a processing algorithm is determined.

  According to the invention of claim 1, since the user can automatically set a necessary inspection program simply by selecting an inspection setting item from the data table, it is not necessary to select an inspection program from an algorithm name or the like. Even a simple user can easily and easily perform desired inspection settings.

  According to the second aspect of the present invention, the user can easily select the setting item in three steps, which are easy to understand the inspection purpose, the feature of the image of the object, and the inspection standard of the object. Setting items for inspection can be selected, and inspection setting work can be performed quickly.

  According to the invention of claim 3, since the setting item can be selected in three easy-to-understand determination steps, even a user unfamiliar with image processing inspection can easily perform inspection setting.

  Hereinafter, an embodiment of the present invention will be described with reference to FIG. The image processing apparatus includes a TV camera 1, an image capture memory 2, a display unit 3, an image display memory 4, an inspection setting unit 5, an overlay memory 6, a control unit 7, a ROM 8, a RAM 9, and an input operation unit 10. . This image processing apparatus is used for measurement, recognition, inspection, and the like of image data features such as shape, area, and position based on image processing of a captured image of an inspection target in an inspection process such as a production line. Is.

  The TV camera 1 images a target object. The image capture memory 2 stores a captured image obtained by capturing an object. The display unit 3 displays a captured image, and a display screen 31 is provided. The image display memory 4 stores a captured image displayed on the display screen 31 of the display unit 3. The inspection setting unit 5 is provided with a data table including a plurality of setting items in advance, and performs a desired inspection setting by a user operation. The overlay memory 6 stores data for overlapping display on the display screen 31 of the display unit 3.

  The control unit 7 includes a CPU, performs calculations on the display image on the display screen 31, performs image processing based on predetermined setting conditions, and controls the entire apparatus. A ROM (storage unit) 8 stores a plurality of inspection programs for inspecting an object and stores an operation procedure of the control unit 7. The RAM 9 has a storage area for the control unit 7 to operate. The input operation unit 10 is, for example, a keyboard, and a predetermined processing condition is input by the user. Further, in the present apparatus, when the image processing operation is performed by the input operation unit 10, an operation guide for setting the inspection process is displayed on the display screen 31 that displays the captured image (see FIG. 4).

  An overall procedure of image processing setting in the image processing apparatus of the present embodiment will be described with reference to FIGS. In the image processing apparatus, as shown in FIG. 2A, first, the user sets the TV camera 1 for imaging the inspection object (S1), and necessary processing for inspecting the inspection object. Algorithm setting is performed (S2). Next, based on the processing algorithm setting, position correction setting for the collation image is performed (S3), numerical values and determination calculation settings necessary for inspection determination are performed (S4), and result output setting for outputting a desired inspection result is set. Perform (S5). In the setting of the processing algorithm in step 2, as shown in FIG. 2B, first, the user selects a processing algorithm (S21), and based on the selected processing algorithm, the inspection of the object is performed. The inspection area desired to be set is set (S22), and the parameters required for the inspection are set (S23). Therefore, in the image processing inspection, the processing algorithm selection (S21) to be performed first is important, and the selection of the processing algorithm affects the optimum and quick processing of the entire image processing inspection. In particular, the image processing apparatus of the present embodiment facilitates this processing algorithm selection.

  Here, in the image processing apparatus of this embodiment, a data table set by the examination setting unit 5 for selecting a processing algorithm will be described with reference to Tables 1 and 2 below.

  In Tables 1 and 2, the setting items for the inspection setting include a plurality of setting items necessary for the inspection setting, items indicating the inspection purpose, items indicating the characteristics of the image of the object, and inspection standards for the object It consists of the item which shows. The items indicating the characteristics of the image of the target object may be “black and white” or “dark”, and the items of the inspection standard correspond to the settings of the characteristics “black and white” and “dark”. Tables 1 and 2 show the request result information P and Q as the result information, respectively. The data tables in Tables 1 and 2 show nine items for inspection purposes in the horizontal direction in the upper column of each table, and the processing algorithm (inspection program) in the vertical direction in the left column. The request result information P and Q are described at the vertical and horizontal matrix positions, respectively. By selecting “inspection purpose”, “feature” and “request result information” from this data table, the vertical processing algorithm in the left column is automatically determined. Table 1 shows that the algorithm to be binarized is used in Table 1, and that the algorithm to be used for density is used in Table 2. In each table, if there is no applicable inspection standard item, it is shaded.

  In the algorithm column of Table 1 above, “line” is an algorithm for detecting the number of objects in a line on the line, and “binarization window” is for binarizing the target image and measuring the area. “Feature extraction” is to measure the number, area, center of gravity, and principal axis direction by binarization in the feature region of the target image, and “binarization edge” is the number by edge detection of the target image on the line, Measure length and coordinates.

  In the algorithm column of Table 2, “Tint Window” measures the average gray value (brightness) of the inspection range, and “Tint Edge” determines the number of edges and the length between edges. is there. “Lead” is used to determine the pitch, width, number, and the like of lead wires such as ICs. “Matching” is to determine the number, the length between coordinates, the coordinate position, the angle, and the area by matching with the registered pattern. Further, “contour matching” is determined by contour matching processing when a plurality of images are overlapped in the target image. Further, “scratch detection” is to detect and determine a place where the brightness changes. Since these processing algorithms are variously known in this type of image processing apparatus, detailed description thereof is omitted here.

  In Tables 1 and 2 above, inspection purposes are usually presence / absence inspection, length inspection, position (deviation) inspection, number (or number) inspection, size inspection, direction inspection, brightness inspection, appearance (dirt, scratch, It is possible to cover almost nine items in the inspection of “burr, chipping” and “difference from standard”. In addition, these items can be understood even by unskilled users.

  For example, the presence / absence inspection is for inspecting whether there is a part, mark, hole, etc. at a specific position of an object. Perform presence / absence inspection. The length inspection is performed to check whether the positional relationship / distance between the two specified points is within the correct range, such as a bolt length check and a dimensional inspection of a molded product. The position (deviation) inspection is to find the position of the specified mark / hole, etc., and inspect whether it is within the correct range. Check the position of the printed circuit board, the label deviation, the screw hole position, etc. Do.

  In the number (or number) inspection, a certain range, the number of marks on a line, the number of IC terminals, and the like are checked. The size inspection is performed based on the size of the area of the object. The direction inspection is performed to check whether the direction of the object is correct, and is used for checking the direction of the IC, the direction of the label, and the like. .

  The brightness inspection is for inspecting the average brightness of the entire inspection range. Appearance inspection includes dirt / scratch inspection and burrs / defects inspection. There are no scratches, dirt, debris, etc. within the specified range, and there are no illegal burrs / burrs on the outline / outer periphery of the object. These are used for checking pinholes and sheet spots. The inspection of the difference from the reference is, for example, an inspection of a difference from the reference pattern or the like such as whether the character pattern is not shifted from the reference pattern on the business card or the like, or there is no unnecessary marking.

  Next, the item indicating the feature of the image of the object represents the image feature as “monochrome” when the captured image is clearly divided in brightness and can be determined in black and white. If the image feature cannot be determined, the image feature is expressed as “shading”. In these selections, when “monochrome” is selected, the processing algorithm can increase the speed of image processing by binarization processing, and when “darkness” is selected, detailed inspection can be performed by darkening processing. .

  Here, the request result information of the inspection standard shown in Tables 1 and 2 will be described. In these tables, the items indicating the inspection standard of the object are specific items required by the user corresponding to the selection of “black and white” and “light / dark” of the characteristics of the image to be inspected for each item of the inspection purpose. Detailed request result information is shown.

  When the inspection purpose is “presence / absence”, whether the inspection target is present or not is selected as a selection item in order to make it easy for the user to understand as the inspection standard request result information. The determination items are “determined by”, “determined by the number of target edges aligned in a row”, “determined by change in brightness”, and “determined by the number of objects that match the registered pattern”. As a result, when the feature of the target image is “black and white”, “determined by the number of objects aligned in a row”, “determined by area”, “determined by change in area or number of objects”, “edges of objects aligned in a row” When the target image feature is “light / dark”, the criteria item of “Judge by number” is “Judge by brightness change”, “Judge by number of target edges in a row”, “Number of items that match registered pattern” "Determine by (when there is no overlap)" and "Determine by the number of items that match the registered pattern" (when there is an overlap).

  When the inspection purpose is “length”, the inspection standard request result information is selected as a selection item for easy understanding by the user, as described above. When the feature selection is “black and white”, “between edges” When the feature selection is “Tint”, “Length between edges”, “Pitch and width of leads”, “Length of target coordinates that match the registered pattern” (when there is no overlap) Each item is provided).

  In addition, when the inspection purpose is position (deviation), as described above, when the feature selection is “black and white”, the items “feature center of gravity coordinate position” and “edge coordinate position” are provided. When the feature selection is “light / dark”, each item of “edge coordinate position” and “target coordinate position that matches the registered pattern” (when there is no overlap, separate when there is) is provided.

  When the inspection purpose is the number (number), the inspection standard request result information includes “number of objects arranged in a row” and “number of objects arranged in a row” when the feature is “black and white”. When each item is set and “Tint” is selected as the feature, each item of “Number of leads” and “Number of objects matching the registered pattern” is set.

  In addition, when the inspection purpose is “size”, the inspection can be performed by measuring the area. Therefore, the inspection standard requirement result information is only used when the feature selection is “black and white”, and “target total area”, “ The item "Individual area of object" is set. This reduces options and simplifies the selection process.

  When the inspection purpose is “direction”, the inspection standard request result information includes only the “target spindle direction” item that can be determined by binarization processing when “black and white” is selected as the feature. When the selection is “light / dark”, only the item of “angle matched with registered pattern” (when there is no overlap, it is divided when there is) is set.

  In addition, when the inspection purpose is “brightness”, the feature selection is only “shading” and the request result information only needs to measure the brightness of the inspection range. Only items.

  When the inspection purpose is “appearance” (dirt, scratches, flashes, shading), the feature selection is only “contrast”, and the appearance can be detected by changes in the brightness of the surroundings such as scratches. Only the item of “detect and judge a change” is set.

  When the inspection purpose is “difference from the reference”, the feature selection is only “light / dark”, and only the item “determined by the area of the portion different from the registered pattern” for detecting the difference from the registered pattern is set.

  As described above, the selection item of the inspection standard is expressed in such a way that the user can easily understand the inspection standard by using the expression of the contents of the request result information that the user wants to know directly. Thereby, even an inexperienced user who does not know the contents of the processing algorithm can easily understand and select the inspection standard items.

  Next, the inspection setting operation procedure in the present embodiment will be described with reference to FIGS. 3, the user selects the inspection purpose shown on the display screen 31 of the display unit 3 in FIG. 4A (here, “a presence / absence” is selected) (S211) (first determination step), Subsequently, when “black and white” of “A” is selected in the characteristics of the image to be inspected (YES in S212) (second determination step), the display screen 31 displays “black and white” in “a presence / absence” of FIG. Instead of the request result information (P) selection screen at the time of selection, the user selects the desired inspection standard request result information (P) from the inspection standards 41 to 44 displayed on this screen (S213) 3 decision steps). By this selection, the control unit 7 automatically calls the processing algorithm (see Table 1) corresponding to one setting item of the inspection criteria 41 to 44 selected at the end of S213 from the ROM 8 (S214), and sets the inspection settings. And the inspection is executed according to the inspection setting.

  On the other hand, when the shading process is selected in S212 (NO in S212) (second determination step), the display screen 31 displays the shading request result information (Q) in FIG. Instead of the selection screen, the user selects a desired inspection standard from the inspection standards 45 to 48 displayed on this screen (S215) (third determination step). By this selection, the control unit 7 automatically calls the processing algorithm (see Table 2) corresponding to one setting item of the inspection criteria 45 to 48 selected at the end of S213 from the ROM 8 (S216), and sets the inspection settings. And the inspection program is executed according to the inspection setting.

  The display in FIGS. 4B and 4C above shows the case where the inspection purpose is the “presence / absence” inspection, but as shown in FIG. 5, it is the same when another inspection purpose is selected. Can be done. That is, by selecting the inspection purpose and features on the display unit 3, the request result information corresponding to these can be displayed on the display screen 31 based on the data tables of Tables 1 and 2, and the inspection setting can be selected. In addition, depending on the inspection purpose, there are cases where there are only options of “black and white” or “light / dark” for the feature selection of the inspection object (for example, direction, brightness, appearance, etc.). Since either one of the processing algorithms of “black and white” or “light / dark” is desirable, the selection is narrowed down in advance.

  As described above, according to the image processing apparatus of the present embodiment, since the necessary inspection program is automatically set according to the setting item last selected by the user, the user can select the setting item from the inspection standard of the data table. You only need to select one. Therefore, the user does not need to select an inspection program such as a specialized algorithm name or application name, and does not have to perform an inspection program selection operation. Therefore, even if the user is unfamiliar with image processing inspection, the desired inspection setting can be set. It can be done easily. In addition, since a data table is provided, selection of setting items can be performed in a small number of selection steps with three steps of inspection purpose, object image characteristics, and inspection criteria, so that a desired selection can be easily and quickly performed. It can be carried out. In addition, by providing an easy-to-understand expression inspection standard in accordance with user request result information in the data table, the user can automatically select a processing algorithm without being aware of the inspection program. Further, the expression of the inspection standard is not an expression limited to a certain industry, so that it can be easily applied to the inspection standard desired by the user and can be applied even to an inspection object of a different industry.

  Further, if a method for inspecting an object using the image processing apparatus is used, even a user unfamiliar with image processing inspection can easily and quickly perform inspection setting.

  In addition, this invention is not restricted to the said embodiment etc., Furthermore, it can change suitably. For example, in this embodiment, the determination step is three steps. However, if necessary, the selection of shades and black and white may be omitted to reduce the number of selection steps, or conversely, the number of selection steps may be increased to three steps or more. It is also possible to perform inspection settings.

1 is a configuration diagram of an image processing apparatus according to an embodiment of the present invention. (A) is a flowchart showing the overall procedure of the image processing apparatus, (b) is a flowchart showing the procedure for setting the processing algorithm of (a). 6 is a flowchart showing a procedure for inspection setting of the image processing apparatus. (A), (b), (c) is a figure which shows the operation guidance screen of each step in the said operation | movement procedure, respectively. The figure which shows the whole outline | summary of the test | inspection setting procedure of the said image processing apparatus.

Explanation of symbols

1 TV camera (imaging part)
5 Inspection setting part 7 Control part 8 ROM (storage part)

Claims (3)

An image processing apparatus that inspects an object based on an image of the object imaged by an imaging unit,
A storage unit storing a plurality of inspection programs;
An inspection setting unit for performing desired inspection settings by a user operation;
A control unit for controlling the entire apparatus,
The inspection setting unit is provided with a data table including a plurality of setting items in advance,
When the inspection setting of the object is performed by the user, after the determination step of selecting one setting item from the plurality of setting items of the data table is performed a plurality of times, the control unit is arranged at the end of the determination step. An image processing apparatus, wherein an inspection program corresponding to a selected setting item is called from the storage unit and inspection is executed. The setting item includes an item indicating an inspection purpose, an item indicating an image characteristic of the object, and an item indicating an inspection standard of the object.
The determination step includes a first determination step of selecting an item indicating the inspection purpose;
A second determining step of selecting an item indicating the feature of the image of the object;
The image processing apparatus according to claim 1, further comprising a third determination step of selecting an item indicating an inspection standard of the object. An image processing method for inspecting an object based on an image of the object imaged by an imaging unit,
When setting up inspection of an object,
A decision step of selecting one item from a plurality of items indicating the inspection purpose;
A determination step of selecting one item from a plurality of items indicating characteristics of the image of the object;
A determination step for selecting one item from a plurality of items, which is an item indicating an inspection standard of an object and whose processing algorithm is determined by being selected;
An image processing method comprising:

Images