CN105044114A - Electrolytic capacitor appearance package defect image detection system and electrolytic capacitor appearance package defect image detection method - Google Patents
Electrolytic capacitor appearance package defect image detection system and electrolytic capacitor appearance package defect image detection method Download PDFInfo
- Publication number
- CN105044114A CN105044114A CN201510206671.8A CN201510206671A CN105044114A CN 105044114 A CN105044114 A CN 105044114A CN 201510206671 A CN201510206671 A CN 201510206671A CN 105044114 A CN105044114 A CN 105044114A
- Authority
- CN
- China
- Prior art keywords
- image
- unit
- detected
- electric capacity
- carries out
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention discloses an electrolytic capacitor appearance package defect image detection system which includes a material feeding trough unit, an upper surface image collecting unit, a lower surface image collecting unit, a side surface image collecting unit, a central processor unit, a display unit and an unqualified product rejection unit. By means of the material feeding trough unit, a vibration disk feeds capacitor materials intermittently according to teeth, wherein the outlet of the vibration disk is a material feeding trough. The vibration disk is a concave disk having spiral lines. The material feeding trough is a transmission transparent trough being limited by two rectangular boards. An electrolytic capacitor detection work station is disposed in the center of the transmission transparent trough. Pins of the electrolytic capacitor are inserted in the lower end of the transmission transparent trough. The transmission transparent trough has three work stations, so that by means of upper surface image collection, lower surface image collection, and side surface image collection, the three directions of the electrolytic capacitor are subjected to image process, thereby recognizing unqualified products having defective appearances.
Description
Technical field
The present invention relates to a kind of electrochemical capacitor appearance packaging defect inspection method, especially based on electrochemical capacitor appearance packaging defect detecting system and the detection method of Digital Image Processing.
Background technology
At present, fault detection for electrochemical capacitor outer packaging also depends on artificial visual and judges, namely, bound edge whether regular by the plastics package circular edge on the upper and lower surface of eye-observation electric capacity whether with electric capacity justified margin, whether have outstanding edge breakage, whether the both positive and negative polarity character position of capacitive surface is correct, whether both positive and negative polarity symbol is corresponding consistent with the length pin of electric capacity, and whether type information prints the defects such as correct.Artificial detection one by one also exists that labour demand is large, speed slow, the problem of inefficiency, drastically influence yield-power and the economic benefit of enterprise.
Summary of the invention
For above-mentioned technical matters, the present invention seeks to, provide a kind of structure simple, manually-operated electric capacity appearance packaging defect automatic checkout system and method can be replaced, improve detection efficiency.
The present invention solves the problems of the technologies described above by the following technical solutions: a kind of electrochemical capacitor appearance packaging defect image detection system, it is characterized in that comprising feed well unit, light source one unit, upper surface image collecting unit, optoelectronic switch one unit, light source two unit, lower surface image collecting unit, light source three unit, optoelectronic switch three unit, side surface image acquisition units, central processor unit, display unit and substandard product culling unit; Described feed well unit carries out interval feeding by tooth to electric capacity by vibrating disk, the outlet of vibrating disk is feed well 1-2, vibrating disk is the concavity dish that a people has helix, feed well is the transmission transparent cell that two rectangular slabs limit, be provided with electrochemical capacitor 5 in the middle of transmission transparent cell and detect station, electrochemical capacitor 6 pin is inserted in transparent cell lower end; Transparent cell is provided with three stations, light source one unit is positioned at directly over the upper surface image collecting unit 2 of station one, for upper surface image collecting unit provides light, upper surface image collecting unit carries out appearance images collection to the upper surface of electric capacity after optoelectronic switch one unit inspection to electric capacity; Light source two unit is positioned at immediately below the lower surface image collecting unit 3 of station two, for lower surface image collecting unit provides light, lower surface image collecting unit carries out appearance images collection to the lower surface of electric capacity after optoelectronic switch two unit inspection to electrochemical capacitor; Light source three unit is positioned at the side of the side surface image acquisition units of station three, for side surface image acquisition units provides light, side surface image acquisition units carries out appearance images continuous acquisition to the side surface of electric capacity after optoelectronic switch three unit inspection to electric capacity, station San Tong method is rotary work station, rotary work station 1-3 drives pin that electric capacity is rotated a circle, trigger when optoelectronic switch three unit inspection enters station three to electric capacity and rotate, the stepper motor swivel plate band dynamic condenser namely on feed well on station rotates a circle.Central processor unit input end receives picture signal and is used for carrying out analyzing and processing to the three-view drawing picture gathered, the display unit display that central processor unit connects gathers image and result, and substandard product culling unit will detect that the defective substandard product of appearance packaging is rejected.
Further, upper and lower surface image carries out coloured image gray processing, image denoising, image enhaucament, region segmentation, image normalization, defects detection respectively, side image carries out coloured image gray processing, image denoising, image enhaucament, image mosaic, region segmentation, image normalization, defects detection respectively, described coloured image gray processing is used for converting the coloured image that video camera obtains to gray level image, described image denoising is used for carrying out noise remove to gray level image, described image enhaucament is used for carrying out contrast strengthen to image, described region segmentation is used for being partitioned into region to be detected, described image normalization is used for rotating image to be detected, the normalized such as cutting and convergent-divergent, described defects detection is used for carrying out defect recognition to normalized image to be detected and existing standard normalized image template, described image mosaic is used for gathering image mosaic to the electric capacity that rotates a circle and becomes a complete side image,
Concrete steps are as follows:
Step 001. optoelectronic switch one detects electric capacity, triggers upper surface image collecting unit camera captures images;
Step 002. converts upper surface coloured image to gray level image;
Step 003. pair upper surface gray level image carries out filtering and removes picture noise;
Upper surface gray level image after step 004. pair denoising carries out image enhaucament to strengthen the contrast of image;
Step 005. carries out region segmentation to the upper surface image after enhancing, specifically comprises step below:
Step 0051. utilizes sobel edge detection operator to carry out rim detection to the upper surface image after enhancing;
Step 0052. pair upper surface image edge utilizes hough change detection electric capacity upper surface outward appearance edge circular curve;
Step 0053. marks and is communicated with circular curve region, electric capacity outward appearance edge, obtains the upper surface region to be detected split;
Step 006. especially, if upper surface packaging has character mark information, also needs to carry out character mark detection, comprises the following steps:
Step 0061. does level or vertical gray-level projection to splitting the upper surface region to be detected obtained, and is partitioned into the specification information region of electric capacity upper surface outward appearance;
The angle that step 0062. utilizes the upper left corner and pixel place, lower left corner straight line in upper surface specification information area image to be detected to depart from vertical direction rotates to vertical direction to the upper surface region to be detected that step 0053 obtains; Wherein top left corner pixel is from left to right, first pixel found from top to bottom, and lower left corner pixel is from left to right, first pixel found from bottom to up.
Step 007. pair cut zone image is normalized, and specifically comprises step below:
Step 0071. is cut into the square-shaped image of rounded edge each 5 pixels up and down to rotating to the upper surface area image to be detected of vertical direction; Be carry out software to any direction image photographed to rotate and make it consistent with standard form image orientation, the concrete anglec of rotation and see step 0062, cutting is also consistent with standard form, in order to not lose any information, doing and expanding among a small circle.
Upper surface area image to be detected after step 0072. pair cutting zooms to the regulation calibration chart picture of size determined pixel point (as 160 × 160);
Standard upper surface template image (its acquisition process is also through step 001-007) in area image to be detected for normalized upper surface and image data base is carried out XOR by step 008., and the judgement that operation result exceedes threshold value is that upper surface is defective.
Step 009. optoelectronic switch two detects electric capacity, triggers lower surface image collecting unit camera captures images;
Step 0010. converts lower surface coloured image to gray level image;
Step 0011. pair lower surface gray level image carries out filtering and removes picture noise;
Lower surface gray level image after step 0012. pair denoising carries out image enhaucament to strengthen the contrast of image;
Step 0013. carries out region segmentation to the lower surface image after enhancing, specifically comprises step below:
Step 00131. utilizes sobel edge detection operator to carry out rim detection to the lower surface image after enhancing;
Step 00132. pair lower surface image edge utilizes hough change detection electric capacity lower surface outward appearance edge circular curve; Step 00133. marks and is communicated with circular curve region, electric capacity outward appearance edge, and obtain the lower surface region to be detected split, mark is here the meaning that image processing field software rectangle circle goes out round edge circle;
Step 0014. especially, if lower surface packaging has character mark information, also needs to carry out character mark detection (also can be used as center or the edge flag in region), it is characterized in that, comprise the following steps:
Step 00141. does level or vertical gray-level projection to splitting the lower surface region to be detected obtained, and is partitioned into the specification information region of electric capacity lower surface outward appearance;
The angle that step 00142. utilizes the upper left corner and pixel place, lower left corner straight line in lower surface specification information area image to be detected to depart from vertical direction rotates to vertical direction to the lower surface region to be detected that step 00133 obtains; Wherein top left corner pixel is from left to right, first pixel found from top to bottom, and lower left corner pixel is from left to right, first pixel found from bottom to up.
Step 0015. pair cut zone image is normalized, and specifically comprises step below:
Step 00151. is cut into the square-shaped image of rounded edge each 5 pixels up and down to rotating to the lower surface area image to be detected of vertical direction;
Lower surface area image to be detected after step 00152. pair cutting zooms to the regulation calibration chart picture of size determined pixel point (as 160 × 160);
Standard lower surface template image (its acquisition process is also through step 009-0015) in area image to be detected for normalized lower surface and image data base is carried out XOR by step 0016., and the judgement that operation result exceedes threshold value is that lower surface is defective.
Step 0017. optoelectronic switch three detects electric capacity, triggers side surface image acquisition units camera captures images, and the swivel plate band dynamic condenser triggered on feed well rotates a circle simultaneously;
Step 0018. converts multiple side surface coloured image to gray level image;
Step 0019. is carried out filtering respectively to multiple side surface gray level image and is removed picture noise;
Multiple side surface gray level images after step 0020. pair denoising carry out image enhaucament respectively to strengthen the contrast of image;
Step 0021. carries out image mosaic to the multiple side surface images after enhancing, specifically comprises the following steps:
The two ends, left and right of the multiple side images of step 00211. after enhancing, respectively extract minutiae;
Step 00212. carries out homotopy mapping to the unique point that two ends, left and right extract;
Step 00213. carries out image co-registration according to match point, obtains spliced view picture side image.
Step 0022. does level or vertical gray-level projection to splicing the view picture side image obtained, and is partitioned into the specification information region of capacitive side appearance;
The angle contralateral surface region to be detected that step 0023. utilizes the upper left corner and pixel place, lower left corner straight line in side surface specification information area image to be detected to depart from vertical direction carries out rotating to vertical direction; Wherein top left corner pixel is from left to right, first pixel found from top to bottom, and lower left corner pixel is from left to right, first pixel found from bottom to up.
Step 0024. pair cut zone image is normalized, and specifically comprises step below:
Step 00241. is cut into the square-shaped image of each 5 pixels up and down apart from edge to rotating to the side surface area image to be detected of vertical direction;
Side surface area image to be detected after step 00242. pair cutting zooms to the regulation calibration chart picture of size determined pixel point (as 160 × 160);
Standard side surface template image (its acquisition process is also through step 0017-0024) in area image to be detected for normalized side surface and image data base is carried out XOR by step 0025., and the judgement that operation result exceedes threshold value is that lower surface is defective.
Described standard upper surface template image adopts step 001 ~ 007 to process generation by specification product to get, standard lower surface template image adopts step 009 ~ 0015 to process generation by specification product to get, and standard Side shuttering image adopts step 0017 ~ 0024 to process generation by specification product to get.
Direction, the electric capacity that described optoelectronic switch detects long pin place can differentiate direction, positive pole symbol place in appearance images in conjunction with being used for.
Described display unit can omnidistance display vision inspection process.
The device that substandard product is rejected by defect classification display unit Presentation Function and different defect classification can be added as required.
The picture number of carrying out image mosaic is chosen in visual field according to side surface collecting unit camera.Described image enhaucament is used for carrying out contrast strengthen to image, described region segmentation is used for being partitioned into region to be detected, described image normalization is used for rotating image to be detected, the normalized such as cutting and convergent-divergent, described defects detection is used for carrying out defect recognition to normalized image to be detected and existing standard normalized image template, and described image mosaic is used for gathering image mosaic to the electric capacity that rotates a circle and becomes a complete side image.
The invention has the beneficial effects as follows, the present invention proposes a kind of electrochemical capacitor appearance packaging defect image detection system and method, adopt above technical scheme compared with existing manual manipulative techniques, there is following Advantageous Effects: application in electric capacity outward appearance detects first, can realize the whether robotization of the situation such as existing defects or mistake of the upper and lower surfactant package edge in capacitor packages packaging and side type information, intellectualized detection, check processing algorithm is effective, speed is fast, verification and measurement ratio advantages of higher.
(1) a kind of electrochemical capacitor appearance packaging defect image detection system of the present invention's design, structure is simple, cost is low, it is easy, multiple functional to realize, and can realize the automatic detection of electric capacity appearance packaging defect, effectively improve detection efficiency.
(2) a kind of electrochemical capacitor appearance packaging defect image detection method of the present invention's design, detect feature for electric capacity outward appearance especially and design, algorithm is effective, under the complicacy condition not increasing algorithm, discrimination is high, and travelling speed is fast, and reliability is high.
Accompanying drawing explanation
Fig. 1 is the functional block diagram of electrochemical capacitor appearance packaging defect image detection system of the present invention;
Fig. 2 is electrochemical capacitor appearance packaging defect image detection method functional block diagram of the present invention;
Fig. 3 is operating diagram.
Embodiment
Below in conjunction with Figure of description, the specific embodiment of the present invention is described in further detail.
As shown in Figure 1,3, the present invention devises a kind of electrochemical capacitor appearance packaging defect image detection system, comprises feed well unit 1, light source one unit, upper surface image collecting unit 3, optoelectronic switch one unit, light source two unit, lower surface image collecting unit 4, light source three unit, optoelectronic switch three unit, side surface image acquisition units 5, central processor unit, display unit and substandard product culling unit.Described feed well unit carries out interval feeding by tooth to electric capacity by vibrating disk 1-1, the outlet of vibrating disk is feed well 1-2, vibrating disk is the concavity dish that a people has helix, electrochemical capacitor can be lined up automatically the downward arrangement of electrochemical capacitor pin, feed well is the transparent cell that two rectangular slabs limit, place electrochemical capacitor 5 in the middle of transparent cell, electrochemical capacitor 6 pin is inserted in transparent cell lower end; Transparent cell is provided with three stations, light source one unit is positioned at directly over the upper surface image collecting unit 2 of station one, for upper surface image collecting unit provides light, upper surface image collecting unit carries out appearance images collection to the upper surface of electric capacity after optoelectronic switch one unit inspection to electric capacity; Light source two unit is positioned at immediately below the lower surface image collecting unit 3 of station two, for lower surface image collecting unit provides light, lower surface image collecting unit carries out appearance images collection to the lower surface of electric capacity after optoelectronic switch two unit inspection to electrochemical capacitor; Light source three unit is positioned at the side of the side surface image acquisition units of station three, for side surface image acquisition units provides light, side surface image acquisition units carries out appearance images continuous acquisition to the side surface of electric capacity after optoelectronic switch three unit inspection to electric capacity, station San Tong method is rotary work station, rotary work station 1-3 drives pin that electric capacity is rotated a circle, trigger when optoelectronic switch three unit inspection enters station three to electric capacity and rotate, the stepper motor swivel plate band dynamic condenser namely on feed well on station rotates a circle.Central processor unit input end receives picture signal and is used for carrying out analyzing and processing to the three-view drawing picture gathered, the display unit display that central processor unit connects gathers image and result, and substandard product culling unit will detect that the defective substandard product of appearance packaging is rejected.Feed well unit has two electrochemical capacitor outlets, and one in the outlet of electrochemical capacitor certified products, and another is unacceptable product outlet, is provided with a mechanical arm and defendant in feed well cell transparent groove is judged that underproof product is allocated to another frame.
Central processor unit is used for carrying out analyzing and processing to the three-view drawing picture gathered, and display unit display gathers image and result, and substandard product culling unit will detect that the defective substandard product of appearance packaging is rejected.
As shown in Figure 2, three directional image that a kind of electrochemical capacitor appearance packaging defect image detection method utilizes upper surface image collection, side surface image acquisition and lower surface image to collect carry out image procossing, identify the defective substandard product of outward appearance.It is characterized in that, upper and lower surface image carries out coloured image gray processing, image denoising (linear filtering method, median filtering method, Wiener Filter Method, Fourier transform and wavelet transformation), image enhaucament (as methods such as histogram, low-pass filter, histogram equalizations), region segmentation, image normalization (Normalized Grey Level combined method), defects detection respectively.Side image carries out coloured image gray processing, image denoising, image enhaucament, image mosaic, region segmentation, image normalization, defects detection respectively.Described coloured image gray processing is used for converting the coloured image that video camera obtains to gray level image, described image denoising is used for carrying out noise remove to gray level image, described image enhaucament is used for carrying out contrast strengthen to image, described region segmentation is used for being partitioned into region to be detected, described image normalization is used for rotating image to be detected, the normalized such as cutting and convergent-divergent, described defects detection is used for carrying out defect recognition to normalized image to be detected and existing standard normalized image template, described image mosaic is used for gathering image mosaic to the electric capacity that rotates a circle and becomes a complete side image.Concrete steps are as follows:
Step 001. optoelectronic switch one detects electric capacity, triggers upper surface image collecting unit camera captures images;
Step 002. converts upper surface coloured image to gray level image;
Step 003. pair upper surface gray level image carries out filtering and removes picture noise;
Upper surface gray level image after step 004. pair denoising carries out image enhaucament to strengthen the contrast of image;
Step 005. carries out region segmentation to the upper surface image after enhancing, specifically comprises step below:
Step 0051. utilizes sobel edge detection operator to carry out rim detection to the upper surface image after enhancing;
Step 0052. pair top surface edge imagery exploitation hough change detection electric capacity upper surface outward appearance edge circular curve;
Step 0053. marks and is communicated with circular curve region, electric capacity outward appearance edge, obtains the upper surface region to be detected split;
Step 006. especially, if upper surface packaging has character mark information, also needs to carry out character mark detection, it is characterized in that, comprise the following steps:
Step 0061. does level or vertical gray-level projection to splitting the upper surface region to be detected obtained, and is partitioned into the specification information region of electric capacity upper surface outward appearance;
The angle that step 0062. utilizes the upper left corner and pixel place, lower left corner straight line in upper surface specification information area image to be detected to depart from vertical direction rotates to vertical direction to the upper surface region to be detected that step 0053 obtains; Wherein top left corner pixel is from left to right, first pixel found from top to bottom, and lower left corner pixel is from left to right, first pixel found from bottom to up.
Step 007. pair cut zone image is normalized, and specifically comprises step below:
Step 0071. is cut into the square-shaped image of rounded edge each 5 pixels up and down to rotating to the upper surface area image to be detected of vertical direction;
Upper surface area image to be detected after step 0072. pair cutting zooms to the regulation calibration chart picture of size determined pixel point (as 160 × 160);
Standard upper surface template image (its acquisition process is also through step 001-007) in area image to be detected for normalized upper surface and image data base is carried out XOR by step 008., and the judgement that operation result exceedes threshold value is that upper surface is defective.
Step 009. optoelectronic switch two detects electric capacity, triggers lower surface image collecting unit camera captures images;
Step 0010. converts lower surface coloured image to gray level image;
Step 0011. pair lower surface gray level image carries out filtering and removes picture noise;
Lower surface gray level image after step 0012. pair denoising carries out image enhaucament to strengthen the contrast of image;
Step 0013. carries out region segmentation to the lower surface image after enhancing, specifically comprises step below:
Step 00131. utilizes sobel edge detection operator to carry out rim detection to the lower surface image after enhancing;
Step 00132. pair lower surface edge imagery exploitation hough change detection electric capacity lower surface outward appearance edge circular curve;
Step 00133. marks and is communicated with circular curve region, electric capacity outward appearance edge, obtains the lower surface region to be detected split;
Step 0014. especially, if lower surface packaging has character mark information, also needs to carry out character mark detection, it is characterized in that, comprise the following steps:
Step 00141. does level or vertical gray-level projection to splitting the lower surface region to be detected obtained, and is partitioned into the specification information region of electric capacity lower surface outward appearance;
The angle that step 00142. utilizes the upper left corner and pixel place, lower left corner straight line in lower surface specification information area image to be detected to depart from vertical direction rotates to vertical direction to the lower surface region to be detected that step 00133 obtains; Wherein top left corner pixel is from left to right, first pixel found from top to bottom, and lower left corner pixel is from left to right, first pixel found from bottom to up.
Step 0015. pair cut zone image is normalized, and specifically comprises step below:
Step 00151. is cut into the square-shaped image of rounded edge each 5 pixels up and down to rotating to the lower surface area image to be detected of vertical direction;
Lower surface area image to be detected after step 00152. pair cutting zooms to the regulation calibration chart picture of size determined pixel point (as 160 × 160);
Standard lower surface template image (its acquisition process is also through step 009-0015) in area image to be detected for normalized lower surface and image data base is carried out XOR by step 0016., and the judgement that operation result exceedes threshold value is that lower surface is defective.
Step 0017. optoelectronic switch three detects electric capacity, triggers side surface image acquisition units camera captures images, and the swivel plate band dynamic condenser triggered on feed well rotates a circle simultaneously;
Step 0018. converts multiple side surface coloured image to gray level image;
Step 0019. is carried out filtering respectively to multiple side surface gray level image and is removed picture noise;
Multiple side surface gray level images after step 0020. pair denoising carry out image enhaucament respectively to strengthen the contrast of image;
Step 0021. carries out image mosaic to the multiple side surface images after enhancing, specifically comprises the following steps:
The two ends, left and right of the multiple side images of step 00211. after enhancing, respectively extract minutiae;
Step 00212. carries out homotopy mapping to the unique point that two ends, left and right extract;
Step 00213. carries out image co-registration according to match point, obtains spliced view picture side image.
Step 0022. does level or vertical gray-level projection to splicing the view picture side image obtained, and is partitioned into the specification information region of capacitive side appearance;
The angle contralateral surface region to be detected that step 0023. utilizes the upper left corner and pixel place, lower left corner straight line in side surface specification information area image to be detected to depart from vertical direction carries out rotating to vertical direction; Wherein top left corner pixel is from left to right, first pixel found from top to bottom, and lower left corner pixel is from left to right, first pixel found from bottom to up.
Step 0024. pair cut zone image is normalized, and specifically comprises step below:
Step 00241. is cut into the square-shaped image of each 5 pixels up and down apart from edge to rotating to the side surface area image to be detected of vertical direction;
Side surface area image to be detected after step 00242. pair cutting zooms to the regulation calibration chart picture of size determined pixel point (as 160 × 160);
Standard side surface template image (its acquisition process is also through step 0017-0024) in area image to be detected for normalized side surface and image data base is carried out XOR by step 0025., and the judgement that operation result exceedes threshold value is that lower surface is defective.
Further, a kind of electrochemical capacitor appearance packaging defect image detection system of the present invention's design and method, described standard upper surface template image adopts step 001 ~ 007 to process generation by specification product to get, and standard lower surface template image adopts step 009 ~ 0015 to process by specification product.
As a kind of optimal technical scheme, direction, the electric capacity that described optoelectronic switch detects long pin place can differentiate direction, positive pole symbol place in appearance images in conjunction with being used for.
As a kind of optimal technical scheme, described display unit can omnidistance display vision inspection process.
As a kind of optimal technical scheme, the device that substandard product is rejected by defect classification display unit Presentation Function and different defect classification can be added as required.
A kind of electrochemical capacitor appearance packaging defect image detection method of the present invention's design, as a kind of optimal technical scheme, the picture number of carrying out image mosaic is chosen in the visual field according to side surface collecting unit camera.
A kind of electrochemical capacitor appearance packaging defect image detection system of the present invention's design and detection method, the lengthy and tedious detection operating process that Traditional Man is visual can be replaced, determine electrochemical capacitor packaging external appearance whether existing defects automatically, it is live that display unit can show the whole differentiation cycle in real time, can also reject according to defect classification (as above surface imperfection, side both positive and negative polarity defect, side model defect information and lower surface defect) classification.Relative to the operation of traditional manual detection, designed system of the present invention and method have that structure is simple, detection efficiency is high, differentiation rate is high, speed is fast, reliability is high, are easy to realize, and with low cost, that applicability is strong advantage.Do not align the edge breakage of electric capacity edge, plastic package at the edge of typical packing of product quality as: plastic package.
To sum up, by setting up and implementing the electrochemical capacitor appearance packaging defect image detection system of view-based access control model imaging device and the method for the present invention's design, can realize to the upper and lower surface packaging edge in capacitor packages packaging, side both positive and negative polarity mark, side type information whether the situation such as existing defects or mistake is carried out high-level efficiency, effectively automatically detected, there is wide market application foreground and economic worth.
By reference to the accompanying drawings embodiments of the present invention are explained in detail above, but the present invention is not limited to above-mentioned embodiment, in the ken that those of ordinary skill in the art possess, can also makes a variety of changes under the prerequisite not departing from present inventive concept.
Claims (7)
1. an electrochemical capacitor appearance packaging defect image detection system, is characterized in that comprising feed well unit, light source one unit, upper surface image collecting unit, optoelectronic switch one unit, light source two unit, lower surface image collecting unit, light source three unit, optoelectronic switch three unit, side surface image acquisition units, central processor unit, display unit and substandard product culling unit; Described feed well unit carries out interval feeding by tooth to electric capacity by vibrating disk, the outlet of vibrating disk is feed well, vibrating disk is the concavity dish that a people has helix, feed well is the transmission transparent cell that two rectangular slabs limit, be provided with electrochemical capacitor in the middle of transmission transparent cell and detect station, electrochemical capacitor pin is inserted in transparent cell lower end; Transparent cell is provided with three stations, light source one unit is positioned at directly over the upper surface image collecting unit 2 of station one, for upper surface image collecting unit provides light, upper surface image collecting unit carries out appearance images collection to the upper surface of electric capacity after optoelectronic switch one unit inspection to electric capacity; Light source two unit is positioned at immediately below the lower surface image collecting unit of station two, for lower surface image collecting unit provides light, lower surface image collecting unit carries out appearance images collection to the lower surface of electric capacity after optoelectronic switch two unit inspection to electrochemical capacitor; Light source three unit is positioned at the side of the side surface image acquisition units of station three, for side surface image acquisition units provides light, side surface image acquisition units carries out appearance images continuous acquisition to the side surface of electric capacity after optoelectronic switch three unit inspection to electric capacity, station San Tong method is rotary work station, rotary work station 1-3 drives pin that electric capacity is rotated a circle, trigger when optoelectronic switch three unit inspection enters station three to electric capacity and rotate, the stepper motor swivel plate band dynamic condenser namely on feed well on station rotates a circle.Central processor unit input end receives picture signal and is used for carrying out analyzing and processing to the three-view drawing picture gathered, the display unit display that central processor unit connects gathers image and result, and substandard product culling unit will detect that the defective substandard product of appearance packaging is rejected.
2. the detection method of electrochemical capacitor appearance packaging defect image detection system according to claim 1, it is characterized in that, the upper and lower surface image of collection carries out coloured image gray processing, image denoising, image enhaucament, region segmentation, image normalization, defects detection respectively, the side image gathered carries out coloured image gray processing, image denoising, image enhaucament, image mosaic, region segmentation, image normalization, defects detection respectively, described coloured image gray processing is used for converting the coloured image that video camera obtains to gray level image, described image denoising is used for carrying out noise remove to gray level image, described image enhaucament is used for carrying out contrast strengthen to image, described region segmentation is used for being partitioned into region to be detected, described image normalization is used for rotating image to be detected, the normalized such as cutting and convergent-divergent, described defects detection is used for carrying out defect recognition to normalized image to be detected and existing standard normalized image template, described image mosaic is used for gathering image mosaic to the electric capacity that rotates a circle and becomes a complete side image,
Concrete steps are as follows:
Step 001. optoelectronic switch one detects electric capacity, triggers upper surface image collecting unit camera captures images;
Step 002. converts upper surface coloured image to gray level image;
Step 003. pair upper surface gray level image carries out filtering and removes picture noise;
Upper surface gray level image after step 004. pair denoising carries out image enhaucament to strengthen the contrast of image;
Step 005. carries out region segmentation to the upper surface image after enhancing, specifically comprises step below:
Step 0051. utilizes sobel edge detection operator to carry out rim detection to the upper surface image after enhancing;
Step 0052. pair upper surface image edge utilizes hough change detection electric capacity upper surface outward appearance edge circular curve;
Step 0053. marks and is communicated with circular curve region, electric capacity outward appearance edge, obtains the upper surface region to be detected split;
If step 006. upper surface packaging has character mark information, also need to carry out character mark detection, comprise the following steps:
Step 0061. does level or vertical gray-level projection to splitting the upper surface region to be detected obtained, and is partitioned into the specification information region of electric capacity upper surface outward appearance;
The angle that step 0062. utilizes the upper left corner and pixel place, lower left corner straight line in upper surface specification information area image to be detected to depart from vertical direction rotates to vertical direction to the upper surface region to be detected that step 0053 obtains; Wherein top left corner pixel is from left to right, first pixel found from top to bottom, and lower left corner pixel is from left to right, first pixel found from bottom to up.
Step 007. pair cut zone image is normalized, and specifically comprises step below:
Step 0071. is cut into the square-shaped image of rounded edge each 5 pixels up and down to rotating to the upper surface area image to be detected of vertical direction;
Upper surface area image to be detected after step 0072. pair cutting zooms to the regulation calibration chart picture of size determined pixel point (as 160 × 160);
Standard upper surface template image (its acquisition process is also through step 001-007) in area image to be detected for normalized upper surface and image data base is carried out XOR by step 008., and the judgement that operation result exceedes threshold value is that upper surface is defective.
Step 009. optoelectronic switch two detects electric capacity, triggers lower surface image collecting unit camera captures images;
Step 0010. converts lower surface coloured image to gray level image;
Step 0011. pair lower surface gray level image carries out filtering and removes picture noise;
Lower surface gray level image after step 0012. pair denoising carries out image enhaucament to strengthen the contrast of image;
Step 0013. carries out region segmentation to the lower surface image after enhancing, specifically comprises step below:
Step 00131. utilizes sobel edge detection operator to carry out rim detection to the lower surface image after enhancing;
Step 00132. pair lower surface image edge utilizes hough change detection electric capacity lower surface outward appearance edge circular curve;
Step 00133. marks and is communicated with circular curve region, electric capacity outward appearance edge, obtains the lower surface region to be detected split;
Step 0014. especially, if lower surface packaging has character mark information, also needs to carry out character mark detection, it is characterized in that, comprise the following steps:
Step 00141. does level or vertical gray-level projection to splitting the lower surface region to be detected obtained, and is partitioned into the specification information region of electric capacity lower surface outward appearance;
The angle that step 00142. utilizes the upper left corner and pixel place, lower left corner straight line in lower surface specification information area image to be detected to depart from vertical direction rotates to vertical direction to the lower surface region to be detected that step 00133 obtains; Wherein top left corner pixel is from left to right, first pixel found from top to bottom, and lower left corner pixel is from left to right, first pixel found from bottom to up.
Step 0015. pair cut zone image is normalized, and specifically comprises step below:
Step 00151. is cut into the square-shaped image of rounded edge each 5 pixels up and down to rotating to the lower surface area image to be detected of vertical direction;
Lower surface area image to be detected after step 00152. pair cutting zooms to the regulation calibration chart picture of size determined pixel point (as 160 × 160);
Standard lower surface template image (its acquisition process is also through step 009-0015) in area image to be detected for normalized lower surface and image data base is carried out XOR by step 0016., and the judgement that operation result exceedes threshold value is that lower surface is defective.
Step 0017. optoelectronic switch three detects electric capacity, triggers side surface image acquisition units camera captures images, and the swivel plate band dynamic condenser triggered on feed well rotates a circle simultaneously;
Step 0018. converts multiple side surface coloured image to gray level image;
Step 0019. is carried out filtering respectively to multiple side surface gray level image and is removed picture noise;
Multiple side surface gray level images after step 0020. pair denoising carry out image enhaucament respectively to strengthen the contrast of image;
Step 0021. carries out image mosaic to the multiple side surface images after enhancing, specifically comprises the following steps:
The two ends, left and right of the multiple side images of step 00211. after enhancing, respectively extract minutiae;
Step 00212. carries out homotopy mapping to the unique point that two ends, left and right extract;
Step 00213. carries out image co-registration according to match point, obtains spliced view picture side image.
Step 0022. does level or vertical gray-level projection to splicing the view picture side image obtained, and is partitioned into the specification information region of capacitive side appearance;
The angle contralateral surface region to be detected that step 0023. utilizes the upper left corner and pixel place, lower left corner straight line in side surface specification information area image to be detected to depart from vertical direction carries out rotating to vertical direction; Wherein top left corner pixel is from left to right, first pixel found from top to bottom, and lower left corner pixel is from left to right, first pixel found from bottom to up.
Step 0024. pair cut zone image is normalized, and specifically comprises step below:
Step 00241. is cut into the square-shaped image of each 5 pixels up and down apart from edge to rotating to the side surface area image to be detected of vertical direction;
Side surface area image to be detected after step 00242. pair cutting zooms to the regulation calibration chart picture of size determined pixel point (as 160 × 160);
Standard side surface template image (its acquisition process is also through step 0017-0024) in area image to be detected for normalized side surface and image data base is carried out XOR by step 0025., and the judgement that operation result exceedes threshold value is that lower surface is defective.
3. a kind of electrochemical capacitor appearance packaging defect image detection system and method according to claim 2, it is characterized in that: described standard upper surface template image adopts step 001 ~ 007 to process generation by specification product to get, standard lower surface template image adopts step 009 ~ 0015 to process generation by specification product to get, and standard Side shuttering image adopts step 0017 ~ 0024 to process generation by specification product to get.
4. a kind of electrochemical capacitor appearance packaging defect image detection system and method according to claim 2, is characterized in that: direction, the electric capacity that described optoelectronic switch detects long pin place can differentiate direction, positive pole symbol place in appearance images in conjunction with being used for.
5. a kind of electrochemical capacitor appearance packaging defect image detection system and method according to claim 2, is characterized in that: described display unit can omnidistance display vision inspection process.
6. a kind of electrochemical capacitor appearance packaging defect image detection system and method according to claim 2, is characterized in that: can add the device that substandard product is rejected by defect classification display unit Presentation Function and different defect classification as required.
7. a kind of electrochemical capacitor appearance packaging defect image detection system and method according to claim 2, is characterized in that: the picture number of carrying out image mosaic is chosen in the visual field according to side surface collecting unit camera.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510206671.8A CN105044114B (en) | 2015-04-27 | 2015-04-27 | A kind of electrolytic capacitor appearance packaging defect image detecting system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510206671.8A CN105044114B (en) | 2015-04-27 | 2015-04-27 | A kind of electrolytic capacitor appearance packaging defect image detecting system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105044114A true CN105044114A (en) | 2015-11-11 |
| CN105044114B CN105044114B (en) | 2018-08-10 |
Family
ID=54450827
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510206671.8A Active CN105044114B (en) | 2015-04-27 | 2015-04-27 | A kind of electrolytic capacitor appearance packaging defect image detecting system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105044114B (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105424714A (en) * | 2015-12-21 | 2016-03-23 | 南京河豚自动化科技有限公司 | Multi-pin-based defect detection device and detection method thereof |
| CN105445286A (en) * | 2015-11-10 | 2016-03-30 | 沈阳建筑大学 | Method and device for detecting quality defects of mark on glass fuse tube products |
| CN106841209A (en) * | 2016-12-28 | 2017-06-13 | 杭州慧芯智能科技有限公司 | One kind is based on big data self study chemical fiber wire ingot intelligence outward appearance detecting system and method |
| CN106932005A (en) * | 2015-12-30 | 2017-07-07 | 希姆通信息技术（上海）有限公司 | Visible detection method and control device that a kind of mobile terminal is tightened up a screw |
| CN107860777A (en) * | 2017-12-20 | 2018-03-30 | 桂林电子科技大学 | A kind of electrolytic capacitor open defect detection means and detection method |
| CN108508023A (en) * | 2018-03-30 | 2018-09-07 | 深圳市益鑫智能科技有限公司 | The defect detecting system of end puller bolt is contacted in a kind of railway contact line |
| CN109799240A (en) * | 2019-02-01 | 2019-05-24 | 信阳舜宇光学有限公司 | A kind of detection method of surface flaw |
| CN109870469A (en) * | 2017-12-05 | 2019-06-11 | 英业达科技有限公司 | The method and system of printed circuit board of the detection with electrolytic capacitor |
| CN109916916A (en) * | 2019-03-29 | 2019-06-21 | 广东利元亨智能装备股份有限公司 | Appearance detecting device |
| CN109978875A (en) * | 2019-04-03 | 2019-07-05 | 无锡立赫智能科技有限公司 | A kind of capacitor open defect recognition methods and identification device |
| CN110160587A (en) * | 2019-06-20 | 2019-08-23 | 珠海格力智能装备有限公司 | Capacitance determining method and device |
| CN110308151A (en) * | 2019-07-22 | 2019-10-08 | 重庆大学 | A kind of surface of workpiece defect identification method and device based on machine vision |
| CN110490955A (en) * | 2019-08-26 | 2019-11-22 | 成都步科智能有限公司 | A kind of mark image acquiring method and device |
| WO2020093293A1 (en) * | 2018-11-06 | 2020-05-14 | 苏州电器科学研究院股份有限公司 | Capacitor detection apparatus and detection method therefor |
| CN111422589A (en) * | 2020-04-07 | 2020-07-17 | 高星东 | Capacitor reverse pole detection direction feeding equipment |
| CN113070228A (en) * | 2021-03-12 | 2021-07-06 | 厦门泓天自动化科技有限公司 | Full-automatic capacitance monitoring machine |
| CN114782429A (en) * | 2022-06-17 | 2022-07-22 | 深圳市菲尼基科技有限公司 | Image-based lithium battery defect detection method, device, equipment and storage medium |
| CN115439452A (en) * | 2022-09-13 | 2022-12-06 | 杭州凯智莆电子有限公司 | Capacitance product detection and evaluation system based on data analysis |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07248300A (en) * | 1994-03-09 | 1995-09-26 | Kyoritsu Denko Kk | Part inspection device |
| JPH0886757A (en) * | 1994-09-16 | 1996-04-02 | Hata Tekkosho:Kk | Device for observing external appearance of powder-compressed molding |
| CN101576510A (en) * | 2008-05-05 | 2009-11-11 | 久元电子股份有限公司 | Detection system and detection method of electronic element |
| CN201373853Y (en) * | 2009-03-06 | 2009-12-30 | 久元电子股份有限公司 | Detecting system for detecting appearances of electronic elements |
| CN101865861A (en) * | 2010-05-28 | 2010-10-20 | 益伸电子(东莞)有限公司 | Automatic detection system for hardware appearance |
| CN202087540U (en) * | 2011-05-19 | 2011-12-28 | 杭州华丰通信器材有限公司 | Online detector of copper piece appearance defects based on machine vision |
| CN102982511A (en) * | 2012-09-17 | 2013-03-20 | 中国人民解放军理工大学气象学院 | Image intelligent optimization processing method |
| CN103264019A (en) * | 2013-04-25 | 2013-08-28 | 洛阳久德轴承模具技术有限公司 | Tapered roller surface defect detection system |
| JP2014106192A (en) * | 2012-11-29 | 2014-06-09 | Hu-Brain Inc | Exterior appearance inspection device |
| CN203720104U (en) * | 2014-01-29 | 2014-07-16 | 洛阳久德轴承模具技术有限公司 | Conveying and image detecting device for bearing roller detecting system |
-
2015
- 2015-04-27 CN CN201510206671.8A patent/CN105044114B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH07248300A (en) * | 1994-03-09 | 1995-09-26 | Kyoritsu Denko Kk | Part inspection device |
| JPH0886757A (en) * | 1994-09-16 | 1996-04-02 | Hata Tekkosho:Kk | Device for observing external appearance of powder-compressed molding |
| CN101576510A (en) * | 2008-05-05 | 2009-11-11 | 久元电子股份有限公司 | Detection system and detection method of electronic element |
| CN201373853Y (en) * | 2009-03-06 | 2009-12-30 | 久元电子股份有限公司 | Detecting system for detecting appearances of electronic elements |
| CN101865861A (en) * | 2010-05-28 | 2010-10-20 | 益伸电子(东莞)有限公司 | Automatic detection system for hardware appearance |
| CN202087540U (en) * | 2011-05-19 | 2011-12-28 | 杭州华丰通信器材有限公司 | Online detector of copper piece appearance defects based on machine vision |
| CN102982511A (en) * | 2012-09-17 | 2013-03-20 | 中国人民解放军理工大学气象学院 | Image intelligent optimization processing method |
| JP2014106192A (en) * | 2012-11-29 | 2014-06-09 | Hu-Brain Inc | Exterior appearance inspection device |
| CN103264019A (en) * | 2013-04-25 | 2013-08-28 | 洛阳久德轴承模具技术有限公司 | Tapered roller surface defect detection system |
| CN203720104U (en) * | 2014-01-29 | 2014-07-16 | 洛阳久德轴承模具技术有限公司 | Conveying and image detecting device for bearing roller detecting system |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105445286A (en) * | 2015-11-10 | 2016-03-30 | 沈阳建筑大学 | Method and device for detecting quality defects of mark on glass fuse tube products |
| CN105424714B (en) * | 2015-12-21 | 2018-04-27 | 南京河豚自动化科技有限公司 | The defects of based on multi-pipe pin detection device and its detection method |
| CN105424714A (en) * | 2015-12-21 | 2016-03-23 | 南京河豚自动化科技有限公司 | Multi-pin-based defect detection device and detection method thereof |
| CN106932005A (en) * | 2015-12-30 | 2017-07-07 | 希姆通信息技术（上海）有限公司 | Visible detection method and control device that a kind of mobile terminal is tightened up a screw |
| CN106841209A (en) * | 2016-12-28 | 2017-06-13 | 杭州慧芯智能科技有限公司 | One kind is based on big data self study chemical fiber wire ingot intelligence outward appearance detecting system and method |
| CN109870469A (en) * | 2017-12-05 | 2019-06-11 | 英业达科技有限公司 | The method and system of printed circuit board of the detection with electrolytic capacitor |
| CN107860777A (en) * | 2017-12-20 | 2018-03-30 | 桂林电子科技大学 | A kind of electrolytic capacitor open defect detection means and detection method |
| CN108508023A (en) * | 2018-03-30 | 2018-09-07 | 深圳市益鑫智能科技有限公司 | The defect detecting system of end puller bolt is contacted in a kind of railway contact line |
| CN108508023B (en) * | 2018-03-30 | 2021-06-04 | 苏州阚创检测有限公司 | Defect detection system for contact end jacking bolt in railway contact network |
| WO2020093293A1 (en) * | 2018-11-06 | 2020-05-14 | 苏州电器科学研究院股份有限公司 | Capacitor detection apparatus and detection method therefor |
| US11402420B2 (en) | 2018-11-06 | 2022-08-02 | Suzhou Apparatus Science Academy Co., Ltd. | Capacitor testing device and testing method thereof |
| CN109799240A (en) * | 2019-02-01 | 2019-05-24 | 信阳舜宇光学有限公司 | A kind of detection method of surface flaw |
| CN109916916A (en) * | 2019-03-29 | 2019-06-21 | 广东利元亨智能装备股份有限公司 | Appearance detecting device |
| CN109978875A (en) * | 2019-04-03 | 2019-07-05 | 无锡立赫智能科技有限公司 | A kind of capacitor open defect recognition methods and identification device |
| CN110160587A (en) * | 2019-06-20 | 2019-08-23 | 珠海格力智能装备有限公司 | Capacitance determining method and device |
| CN110160587B (en) * | 2019-06-20 | 2024-01-30 | 珠海格力智能装备有限公司 | Capacitance detection method and system |
| CN110308151A (en) * | 2019-07-22 | 2019-10-08 | 重庆大学 | A kind of surface of workpiece defect identification method and device based on machine vision |
| CN110490955A (en) * | 2019-08-26 | 2019-11-22 | 成都步科智能有限公司 | A kind of mark image acquiring method and device |
| CN111422589A (en) * | 2020-04-07 | 2020-07-17 | 高星东 | Capacitor reverse pole detection direction feeding equipment |
| CN113070228A (en) * | 2021-03-12 | 2021-07-06 | 厦门泓天自动化科技有限公司 | Full-automatic capacitance monitoring machine |
| CN114782429A (en) * | 2022-06-17 | 2022-07-22 | 深圳市菲尼基科技有限公司 | Image-based lithium battery defect detection method, device, equipment and storage medium |
| CN115439452A (en) * | 2022-09-13 | 2022-12-06 | 杭州凯智莆电子有限公司 | Capacitance product detection and evaluation system based on data analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105044114B (en) | 2018-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105044114A (en) | Electrolytic capacitor appearance package defect image detection system and electrolytic capacitor appearance package defect image detection method | |
| CN204789357U (en) | Electrolytic capacitor outward appearance packing defect image detection system | |
| CN105181714B (en) | A kind of bottling liquid Impurity Detection System and its detection method based on image procossing | |
| CN109034165B (en) | Method, device and system for cutting certificate image and storage medium | |
| CN107203990A (en) | A kind of labeling damage testing method based on template matches and image quality measure | |
| CN108982512A (en) | A kind of circuit board detecting system and method based on machine vision | |
| CN104766078B (en) | The inclination angle detection method of target image and inclination angle detection device | |
| US10586321B2 (en) | Automatic detection, counting, and measurement of lumber boards using a handheld device | |
| CN105139386A (en) | Image processing method for quickly and automatically detecting soldered dot unsatisfied products of electric connectors | |
| CN105426894B (en) | Railway plug pin image detecting method and device | |
| CN102507598A (en) | High-speed unordered capsule defect detecting system | |
| CN104597057A (en) | Columnar diode surface defect detection device based on machine vision | |
| CN104297255A (en) | Visual inspection method and system device for paper cup defects | |
| US20100079453A1 (en) | 3D Depth Generation by Vanishing Line Detection | |
| CN106469312A (en) | Weld seam visual identity method based on region growing label | |
| CN104021382A (en) | Eye image collection method and system | |
| CN104034638A (en) | Diamond wire particle online quality inspection method based on machine vision | |
| CN107490583A (en) | A kind of intermediate plate defect inspection method based on machine vision | |
| CN110032946A (en) | A kind of aluminium based on machine vision/aluminium blister package tablet identification and localization method | |
| CN109146841A (en) | A kind of visible detection method and system of dry cell pulp layer paper filling defect | |
| CN106780437B (en) | A kind of quick QFN chip plastic packaging image obtains and amplification method | |
| CN103245666B (en) | Automatic detecting method for appearance defects of storage battery polar plate | |
| CN109378279A (en) | Wafer detection method and wafer detection system | |
| WO2023168984A1 (en) | Area-array camera-based quality inspection method and system for cathode copper | |
| CN205721986U (en) | A kind of cable visual defects detection and character recognition device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190717 Address after: Nanjing City, Jiangsu province 210007 Yudaojie Camp No. 2 Box 32 (Institute of Communication Engineering) Patentee after: National University of Defense Technology Address before: Nanjing City, Jiangsu province 210007 Yudaojie Camp No. 2 Box 32 (Institute of Communication Engineering) Patentee before: PLA University OF SCIENCE AND TECHNOLOGY |