JP2008039479A - Method and apparatus for inspecting water leakage of breathable membrane, and method for manufacturing breathable member - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method and apparatus for inspecting water leakage of a breathable membrane, which judges presence or absence of water leakage accurately and promptly, unaffected by a change in shape or brightness of the breathable membrane itself. <P>SOLUTION: The inspection method related to the present invention comprises: a process A in which hydraulic pressure is applied from an internal bottom face 18p side to the breathable membrane 22; a process B in which an image of the internal bottom face 18p is acquired by the breathable membrane 22 in chronological order; and a process C in which presence or absence of water leakage from an external bottom face 18q side to the internal bottom face 18p side is judged, based on the difference image between an image last acquired and an image presently acquired. The acquisition of an image by the process B and the judgment by process C are alternately performed so that the judgement is performed k times for (k+1) times of acquisitions of the images (where k is an integer of two or more) while hydraulic pressure is kept applied to the breathable membrane 22. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a method for inspecting the presence or absence of water leakage in a gas permeable membrane, and an apparatus that can be suitably used for the method. The present invention further relates to a method for manufacturing a ventilation member including an inspection step using the method.

  There are ventilation members attached to an ECU box or a headlight housing of an automobile as a target for performing a water leak inspection. A representative example of the ventilation member is shown in FIG. 7A. The ventilation member 18 includes a cylindrical main body 20 and a ventilation film 22 attached to the main body 20. The water leakage inspection of the ventilation member 18 is an inspection in which water pressure is applied to the ventilation film 22 from the outer bottom surface 18q side to check whether water leakage occurs on the inner bottom surface 18p side.

As a method of water leakage inspection, there are a visual method and an image processing method, but in recent years, an image processing method has become mainstream. For example, Patent Document 1 describes an inspection method that utilizes the fact that water absorbs infrared rays. Patent Document 2 describes an inspection method that can identify the type of leakage by using a visible image and an infrared image.
Japanese Patent No. 3333132 JP-A-6-331480

  By the way, during the water leakage inspection of the ventilation member 18 illustrated in FIG. 7A, the shape and brightness of the ventilation film 22 may gradually change due to the application of water pressure. Such a change usually occurs even when there is no water leak. Therefore, when adopting an inspection method based on image processing, it is important to take measures so that the above change is not erroneously judged as NG.

  The method described in Patent Document 1 determines the presence or absence of water leakage based on the infrared reflection intensity distribution when there is no water leakage. Therefore, no solution can be provided for the problem that a change in the gas permeable membrane itself causes an erroneous determination.

  Patent Document 2 (particularly, paragraph 0008, FIG. 4 and FIG. 5) describes that the presence or absence of leakage is determined based on the area value of a binary image obtained by binarizing the difference image. Yes. As shown in the time chart of FIG. 8, the method described in Patent Document 2 accumulates binary images obtained from difference images, and uses the accumulated binary images to determine whether or not there is leakage. The type is specified. This method has a problem that the occurrence of water leakage cannot be detected during the accumulation of binary images, although the influence of changes in the shape of the gas permeable membrane and the luminance over time hardly affects the determination result.

  Accordingly, an object of the present invention is to provide a method and an apparatus for inspecting water leaks in a gas permeable membrane, which can detect the occurrence of water leaks quickly and accurately regardless of changes in the shape and brightness of the gas permeable membrane itself. And Moreover, the manufacturing method of the ventilation member including the test | inspection process using the method is provided.

That is, the present invention
Step A for applying water pressure from the first main surface side to the gas permeable membrane that prohibits the passage of water and allows the passage of gas;
Step B for acquiring an image of the second main surface of the ventilation membrane in time series,
A step C of determining the presence or absence of water leakage from the first main surface side to the second main surface side based on a difference image between the most recently acquired image and the currently acquired image,
The acquisition of the image by the process B and the determination by the process C are alternately performed so that the determination is executed k times for the (k + 1) times of image acquisition while continuing the state in which the water pressure is applied to the ventilation membrane. Provide a water leak inspection method. (However, k is an integer of 2 or more.)

The present invention also provides:
An inspection jig that prohibits the passage of water and supports a gas permeable membrane that allows the passage of gas, and that applies water pressure from the first main surface side;
A camera that is disposed at a position facing the gas permeable membrane supported by the inspection jig and images the second main surface of the gas permeable membrane;
The second main surface from the first main surface side based on the difference image between the means for acquiring the image of the second main surface of the air permeable membrane in time series from the camera, the image acquired in the past in the past, and the currently acquired image An image processing apparatus including means for determining the presence or absence of water leakage to the side,
With
A water leak inspection apparatus that alternately executes image acquisition and determination so that determination is executed k times for (k + 1) image acquisition while continuing the state in which water pressure is applied to the ventilation membrane. I will provide a. (However, k is an integer of 2 or more.)

Furthermore, the present invention provides:
Producing a ventilation member including a cylindrical main body having openings at both ends, and a ventilation film attached to the main body to form a bottom surface;
Inspecting the ventilation membrane attached to the ventilation member as an inspection target by the method described above,
A process of sorting the ventilation member into a non-defective product and a defective product based on the result of the inspection;
A method for manufacturing a ventilation member is provided.

  According to the inspection method, in parallel with the process B of imaging the second main surface of the gas permeable membrane in time series, the first image is obtained based on the difference image between the most recently acquired image and the currently acquired image. Step C for determining the presence or absence of water leakage from the main surface side to the second main surface side is repeated a plurality of times. The difference image between the most recently acquired image and the currently acquired image is that the two images that are the basis of this difference image are close in time. It is difficult to include information. In other words, by creating difference images in time series and repeatedly performing the determination using these difference images several times along the time axis, it is not affected by minute shape changes or brightness changes of the ventilation membrane itself. It becomes possible to detect water leaking from the gas permeable membrane. Furthermore, since the determination itself is also performed in time series, it is possible to immediately detect the occurrence of water leakage. Further, by setting the number of image acquisition times (number of imaging times) to (k + 1) times and the number of times of creating and determining a difference image to be k times, the burden on the computer that performs image processing can be reduced to the maximum.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

  FIG. 1 is a schematic diagram showing an outline of a water leak inspection apparatus for a gas permeable membrane. An example of an inspection object is a ventilation member 18 attached to an ECU box or a headlight housing of an automobile. The water leak inspection apparatus 100 includes an inspection jig 16 for attaching the ventilation member 18, an illumination 15 for irradiating light toward the ventilation member 18 attached to the inspection jig 16, and a ventilation attached to the inspection jig 16. The camera 12 which images the internal bottom face 18p of the member 18 and the image processing apparatus 38 which processes the image acquired from the camera 12 and determines the presence or absence of water leak are provided.

  As shown in FIG. 7A, the ventilation member 18 includes a cylindrical main body 20 having openings at both ends, and a ventilation film 22 attached to one end of the main body 20 to form an inner bottom surface 18p. The ventilation member 18 to which the gas permeable membrane 22 is attached has a bottom portion made of the gas permeable membrane 22 at one end, and the inner bottom surface 18p is visible from the opening at the other end.

  The main body 20 of the ventilation member 18 is a cylindrical part made of, for example, a resin having rubber elasticity. The gas permeable membrane 22 is a resin porous membrane having a function of prohibiting the passage of water and allowing the passage of gas, and is fixed to the main body 20 by heat welding or using an adhesive. Whether or not the gas permeable membrane 22 is firmly attached to the main body 20, in other words, whether or not water enters the main body 20 even when water pressure is applied to the gas permeable membrane 22 is inspected by image processing.

  The inspection jig 16 can be used to perform imaging by the camera 12 while applying water pressure to the gas permeable membrane 22. The inspection jig 16 has a chamber 16h capable of containing water therein, and has an opening 161 for mounting the ventilation member 18 and a water injection port 162 for injecting water W into the chamber 16h. It is provided above and below the chamber 16h.

  As shown in FIG. 1, the ventilation member 18 is attached to the opening 161 of the inspection jig 16 such that the ventilation film 22 is exposed inside the chamber 16h and the opening on the opposite side is exposed outside the chamber 16h. The opening 161 is provided with a seal ring 163 that fills the gap between the inspection jig 16 and the ventilation member 18 so that water W in the chamber 16h does not leak from the gap. After the ventilation member 18 is attached to the opening 161, water W is injected into the chamber 16h from the water injection port 162, whereby an appropriate water pressure is applied to the ventilation film 22 of the ventilation member 18. When the air permeable membrane 22 is damaged or the adhesion to the main body 20 is insufficient, the water W leaks inward, and the leak is caught by the camera 12.

  An image serving as a determination material for the presence or absence of water leakage is acquired from the camera 12. The camera 12 that images the region where the ventilation member 18 is arranged includes, for example, a two-dimensional CCD as an imaging element, and is prepared at a position facing the ventilation member 18 attached to the inspection jig 16. Specifically, the camera 12 is installed directly above the ventilation member 18. As a result, the entire inner bottom surface 18p of the ventilation member 18 can be stored in the image.

  The illumination 15 is, for example, a fluorescent lamp, a halogen lamp, a metal halide lamp, an LED lamp, or the like, and is arranged to irradiate light (visible light and / or infrared light) from an oblique direction to the inner bottom surface 18p of the ventilation member 18. Yes. The type of the illumination 15 can be selected from the above according to the type of the ventilation film 22 to be imaged. Further, an arrangement in which light is irradiated from a direction substantially parallel to a line connecting the center of the ventilation member 18 and the center of the camera 12 may be employed.

  Next, FIG. 2 is a block diagram of the water leak inspection apparatus 100 shown in FIG. The water leak inspection apparatus 100 includes a camera 12, an image processing module 30, an analysis computer 32, and a monitor 34. The image processing module 30 is connected to the PCI bridge of the analysis computer 32 and transfers an image acquired from the camera 12 to the analysis computer 32. The analysis computer 32 analyzes the image transferred from the image processing module 30 and determines the presence or absence of water leakage. The analysis computer 32 also serves as a controller for the pump 36 for injecting water into the inspection jig 16 (see FIG. 3). The monitor 34 functions as a display unit that outputs the determination result of the analysis computer 32. The image processing module 30, the analysis computer 32, and the monitor 34 constitute an image processing device 38.

  FIG. 3 is a flowchart of an inspection program executed by the analysis computer. When this inspection program is started, water is first poured into the chamber 16h of the inspection jig 16 (step S1). Thereby, a predetermined water pressure is applied to the ventilation film 22 of the ventilation member 18. The state where water pressure is applied to the gas permeable membrane 22 is maintained for a certain period of time. In the meantime, it is checked whether or not there is a water leak by the processing in step S2 and subsequent steps.

  When water injection is completed, an imaging command is given to the camera 12. An image captured by the camera 12 is stored in the memory of the image processing module 30. The analysis computer 32 acquires the images stored in the memory of the image processing module 30 as a first set (n = 1) of images and stores them in its own main memory (step S2). Next, while waiting for a predetermined time (for example, several seconds), a counter for measuring the set number n is incremented (steps S3 and S4).

  After waiting for a predetermined time Tn, the second set (n = 2) of images are acquired in the same procedure as the first set of images (step S5). In this manner, the ventilation member 18 is imaged by the camera 12 in a predetermined time schedule while maintaining the relative positional relationship between the ventilation member 18 and the camera 12. The standby time Tn may be constant or may be different for each set of images.

  Next, the analysis computer 32 calculates a difference image between the first set (n = 1) of images stored in the main memory and the second set (n = 2) of images stored in the main memory. Is generated (step S6). The generated difference image is binarized with a predetermined threshold value (step S7). The presence or absence of water leakage is determined from the binarized image thus obtained (step S8). That is, based on the difference image between the most recently acquired image and the currently acquired image, it is determined whether or not a defect (water leakage) has occurred on the inner bottom surface 18p. The criterion for determining that there is water leakage is, for example, that the area or size of the binarized image exceeds a predetermined value.

  If it is determined in step S8 that there is a water leak, an eye sage indicating that a water leak has occurred is displayed on the monitor 34 (step S10). Further, in step S9, it is confirmed whether or not the predetermined number of determinations has been reached. If not, the processing from step S3 is repeated. The number of determinations can be set to about 10 times, for example. Instead of the number of determinations, whether or not a predetermined inspection time has passed may be used as a condition for the inspection end.

  During the inspection, the shape and brightness of the gas permeable membrane 22 may gradually change due to application of water pressure. Such a change may occur even when there is no water leakage, and it is necessary to devise a method for preventing such a change from being erroneously determined as NG. According to the present embodiment shown in the flowchart of FIG. 3, a difference image between the n-th set of images and the (n + 1) -th set of images acquired in time series is generated, and the generated difference image is binarized with an appropriate threshold value. The presence or absence of water leakage is determined using the binarized image obtained by the conversion.

  That is, as shown in the time chart of FIG. 3, an image of the inner bottom surface 18p (second main surface) by the gas permeable membrane 22 at a predetermined time interval while continuing the state in which the water pressure is applied to the gas permeable membrane 22 of the gas permeable member 18. Is acquired from the camera 12, and in parallel with this process, based on the difference image between the most recently acquired image and the currently acquired image, from the external bottom surface 18q side to the internal bottom surface 18p side. The process of determining the presence or absence of water leakage is repeated a plurality of times.

  In this way, images are acquired continuously (in time series) at predetermined times, and differences between these images are found. When such a method is employed, the deformation of the gas permeable membrane 22 hardly affects the determination result, and the occurrence probability of erroneous determination can be considerably reduced. In addition, since the difference images are generated in time series and sequentially determined, even if water leaks suddenly, it is possible to make an NG determination immediately without overlooking this. In addition, according to the method of the present embodiment shown in the flowchart, the determination is performed k times during (k + 1) times of imaging, which is very efficient, and also from the viewpoint of reducing the load on the image processing device 38. Preferred (where k is an integer of 2 or more).

  Further, as shown in FIG. 5, even if the imaging target has a complicated shape, a difference image between the past image 40 and the current image 41 is generated, and the difference image is binarized. Thus, it becomes possible to easily extract the defect 42 (water leakage).

  In this embodiment, the water pressure applied to the gas permeable membrane 22 is constant from the start to the end of the inspection, but the water pressure applied to the gas permeable membrane 22 changes with time by gradually increasing the amount of water injected into the chamber 16h. (Specifically, it gradually increases). In the present embodiment, the water leak determination is repeatedly performed until the number of determinations reaches the specified number. However, the inspection may be stopped when the water leak is detected. Specifically, the execution of the inspection program shown in the flowchart of FIG. 3 is stopped, and the water in the chamber 16h of the inspection jig 16 is drained. By doing so, it is possible to prevent water from overflowing to the surroundings and from splashing the apparatus or getting the work area wet with water.

Further, in order to further reduce the influence of the change of the gas permeable membrane 22 itself during the period of applying the water pressure on the determination result, the following measures (1) and (2) can be taken.
(1) The process from step S2 in the flowchart of FIG. 3 is started on the condition that a predetermined waiting time elapses after the water pressure is started.
(2) The image acquisition time interval is shortened in an initial stage from when the water pressure starts to be applied until a predetermined time elapses, and after the predetermined time has elapsed, the image acquisition time interval is increased.

  In order to perform a more accurate inspection, it is desirable to illuminate so that no shadow is generated inside the ventilation member 18 to be imaged. For example, as shown in FIG. 6, according to the coaxial epi-illumination using the half mirror 24, such a shadow does not occur inside the ventilation member 18, which is suitable for the method of the present invention.

  When applying the inspection method of this embodiment to the manufacturing process of the ventilation member 18, the process of manufacturing the ventilation member 18 is performed first. Next, a water leak inspection is performed by the method of the present embodiment with the ventilation member 18 including the ventilation film 22 as an inspection target (inspection process). Based on the result of the inspection process, the ventilation member 18 is sorted into a non-defective product and a defective product (sorting process).

  The ventilation member 18 selected as a non-defective product may be further subjected to a process of attaching a cover 21 that protects the ventilation film 22 as in the ventilation member 18B shown in FIG. 7B. The cover 21 is attached only to the ventilation member 18 selected as a non-defective product without water leakage in the inspection process. When viewed from the ventilation member 18B of FIG. 7B, the ventilation member 18 of FIG. 7A is an intermediate product, and therefore, the cover 21 is not wasted by determining pass / fail at the time of the intermediate product.

  The inspection method according to the present invention can be suitably applied to inspection of a single gas permeable membrane and products using a gas permeable membrane other than the gas permeable member.

The schematic diagram which shows the outline | summary of the inspection method concerning this invention Block diagram of inspection equipment Flow chart of inspection program Time chart corresponding to the flowchart of FIG. Image example for NG determination Schematic diagram showing preferred arrangement of light sources Cross-sectional view of the ventilation member to be inspected Sectional drawing of other examples of ventilation member Time chart showing an overview of conventional inspection methods

Explanation of symbols

12 Camera 16 Inspection jig 18 Ventilation member 20 Main body 22 Venting membrane 18p Internal bottom surface 18q External bottom surface 30 Image processing module 32 Analysis computer 34 Monitor 38 Image processing device 100 Water leak inspection device W Water

Claims (4)

Step A for applying water pressure from the first main surface side to the gas permeable membrane that prohibits the passage of water and allows the passage of gas;
Step B for acquiring an image of the second main surface of the gas permeable membrane in time series;
A step C of determining the presence or absence of water leakage from the first main surface side to the second main surface side based on a difference image between the image acquired in the latest past and the currently acquired image; Including
The acquisition of the image by the process B and the process C are performed so that the determination is performed k times with respect to (k + 1) acquisitions of the image while continuing the state in which the water pressure is applied to the ventilation membrane. A water leakage inspection method that alternately executes the determination by the method. (However, k is an integer of 2 or more.)   The water leakage according to claim 1, wherein in step C, it is determined that there is water leakage when the area of the binarized image obtained by binarizing the difference image with a predetermined threshold value is greater than or equal to a predetermined value. Inspection method. An inspection jig that prohibits the passage of water and supports a gas permeable membrane that allows the passage of gas, and that applies water pressure from the first main surface side;
A camera that is disposed at a position facing the gas permeable membrane supported by the inspection jig and that images the second main surface of the gas permeable membrane;
The first main surface side based on a difference image between the image acquired on the second main surface of the gas permeable membrane in time series from the camera, the image acquired in the last past, and the image acquired at present And an image processing device including means for determining the presence or absence of water leakage from the second main surface side,
The acquisition of the image and the determination are alternately performed so that the determination is performed k times for the (k + 1) acquisition of the image while continuing the state in which the water pressure is applied to the ventilation membrane. Water leak inspection device. (However, k is an integer of 2 or more.) Producing a ventilation member including a cylindrical main body having openings at both ends, and a ventilation film attached to the main body to form a bottom surface;
Inspecting the ventilation membrane attached to the ventilation member as an inspection target by the method according to claim 1 or 2,
A step of sorting the ventilation member into a non-defective product and a defective product based on the result of the inspection;
A method for manufacturing a ventilation member.

Images