CN112730458A - Automatic television backboard appearance detection device and detection method - Google Patents
Automatic television backboard appearance detection device and detection method Download PDFInfo
- Publication number
- CN112730458A CN112730458A CN202110113301.5A CN202110113301A CN112730458A CN 112730458 A CN112730458 A CN 112730458A CN 202110113301 A CN202110113301 A CN 202110113301A CN 112730458 A CN112730458 A CN 112730458A
- Authority
- CN
- China
- Prior art keywords
- detection
- camera
- line
- frame
- sliding
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
Abstract
The utility model provides a TV backplate mesoscopic detection device, includes the frame, and the first half of this frame is equipped with the detection case, and the lower half of this frame is formed with the transmission space that supplies the assembly line to pass, be equipped with the slide bar on the horizontal direction in the upper portion of detection case, be equipped with the camera installation mechanism of two left and right interval arrangements on this slide bar, this two-phase machine installation mechanism can set up on the slide bar with sliding, two camera installation mechanisms all are equipped with the line and sweep the camera, the bottom of detection case is hollow out construction, the line is swept the camera and is set up with liftable, one side of frame is equipped with the inside electric cabinet that has control system, and a side of this frame is equipped with the display screen, be equipped with photoelectric sensing ware and encoder in the output space, camera, display screen, photoelectric sensing ware and encoder are all connected with control system signal to. The invention can realize the automatic detection of the television backboard, greatly improves the detection efficiency, and can be directly matched with the existing production line for use without interrupting the production line.
Description
Technical Field
The invention relates to a detection device and a detection method, in particular to a television backboard appearance detection device and a detection method.
Background
For television production, a plurality of stations on a production line are assembled, and a television set seen on the market can be obtained only through a back plate, one LED lamp, a screw, an outer frame, a circuit board, a rear shell and the like and a final detection test. The method aims at detecting the inner appearance of the television backboard by adopting manual detection, and is difficult to avoid visual fatigue and has the risk of missed detection in the face of numerous detection items. And secondly, the manual detection causes the conditions of low efficiency, high missing detection risk and the like under the environment of rapid industrial development.
The industry is still not ripe automatic checkout equipment to detecting the TV set backplate, and only few equipment limitations are also very big, or detect that the thing is single, can't satisfy multiple detection demand, or can't satisfy all detectable of size in size. Moreover, the existing automatic detection equipment needs to break the original production line, and then the automatic detection equipment is added at the breaking position.
Disclosure of Invention
The invention provides a television backboard appearance detection device and a television backboard appearance detection method, and mainly aims to overcome the problems in the prior art.
The invention adopts the following technical scheme:
the utility model provides a TV backplate mesoscopic detection device, includes the frame, and the first half of this frame is equipped with the detection case, and the lower half of this frame is formed with the transmission space that supplies the assembly line to pass, be equipped with the slide bar on the horizontal direction in the upper portion of detection case, be equipped with the camera installation mechanism of two left and right interval arrangements on this slide bar, this two-phase machine installation mechanism can set up on the slide bar with sliding, two camera installation mechanisms all are equipped with the line and sweep the camera, the bottom of detection case is hollow out construction, the line is swept the camera and is set up with liftable, one side of frame is equipped with the inside electric cabinet that has control system, and a side of this frame is equipped with the display screen, be equipped with photoelectric sensing ware and encoder in the output space, camera, display screen, photoelectric sensing ware and encoder are all connected with control system signal to.
Further, camera installation mechanism includes riser, slider, slide and camera mounting bracket, the rear end face of riser is located to the slider, the preceding terminal surface and the liftable ground setting of riser are located to the slide, the preceding terminal surface of slide is located to the camera mounting bracket, the top and the camera mounting bracket of camera are swept to the line are connected, the slider is passed to the slide bar.
Furthermore, the number of the slide rods is two, the two slide rods are arranged at intervals up and down, two slide blocks are arranged on each camera mounting mechanism, and each slide block corresponds to one slide rod.
Furthermore, the left and right side faces of the vertical plate are respectively provided with a left sliding groove in a shape of ">" and a right sliding groove in a shape of "<", the rear end of the sliding plate is provided with an accommodating groove, the left side of the accommodating groove is in a shape of "/", the right side of the accommodating groove is in a shape of "\\", the front end of the vertical plate is matched with the accommodating groove, the front end of the vertical plate is embraced by the accommodating groove, the left and right ends of the sliding plate are respectively positioned in the left sliding groove and the right sliding groove, the middle of the vertical plate is provided with a vertical screw rod groove, a screw rod is arranged in the screw rod groove in a penetrating mode, the bottom of the vertical plate is penetrated by the bottom end of the screw rod and connected with a hand wheel, the rear.
A method for automatically detecting the inner appearance of a television backboard comprises the following steps:
1) the photoelectric sensor is used for sensing information of a television backboard on the water flow plate, when the control system receives shielding information of the water flow plate sensed by the photoelectric sensor, the two line-scan cameras are triggered to fetch images according to pulse frequency given by the encoder, when the water flow plate reaches the photoelectric sensor at the lower station and cannot sense the shielding information of the water flow plate, the image fetching of the two line-scan cameras is finished, the control system carries out picture arrangement processing on the images fetched by the two line-scan cameras, all the images fetched at this time are combined into one large image, and the large image is displayed on the display screen;
2) the control system creates a detection formula template of items to be detected of the television backboard, the detection formula template contains a type formula of each item to be detected, then six positioning points are selected on a spliced big picture and are learned, after the learning of the positioning points is completed, the items to be detected are continuously learned, the learning of the detection items is to select partial frames to be detected, select corresponding type formulas, and after all the items to be detected are learned, the detection formula template is stored;
3) and (3) repeating the step 1), starting to detect the television back plates of the same type, displaying the detection result on a display screen, respectively representing the correctness and the mistake of the detection result by using frame bodies with different colors, storing the correct picture into a designated folder after the detection is finished, storing the wrong picture into another folder, and counting all the items detected each time to facilitate the query.
Further, the detection work in the step 3) includes positioning and framing on the large graph, the positioning work is to adopt one line scan camera to select three positioning points, the two line scan cameras have six positioning points for affine transformation, a transformation matrix of the line scan camera is obtained through the positioning points, and then each detection item position is transformed through an inverse matrix to obtain a corresponding detection item position on the large graph during detection; and the box selector is used for performing box selection on the detection item position on the large graph, when the detection item position is selected in a box mode, a corresponding type formula is selected, and after the type formula is selected, the control system automatically analyzes the detection result.
Further, the detection items comprise black screws, small black screws, black screw black backgrounds, white screws, white screw black backgrounds, multiple screw locks, wire arranging adhesive tapes, light reflecting adhesive tapes, small adhesive tapes, adhesive tape breakage and tilting, adhesive tape multiple sticking, label deviation, FFC, flexible flat cables, Pin, small Pin, buckles, piano covers, lamp holders, Boss columns, wire breakage, mirror surface white screws and detection of the existence of labels.
As can be seen from the above description of the present invention, compared with the prior art, the present invention has the following advantages:
1. the device can be directly placed on the existing production line, and the production line can pass through the transmission space without breaking the existing production line. A
2. The device and the method can simultaneously detect various tested components and can test the sizes of the tested components. The method realizes one-time modeling and multiple uses, can realize immediate switching and immediate detection on the established template formula, and has great improvement on efficiency. Relative artifical the detection, not only the promotion of efficiency, including lou examining the risk and all having very big reduction, still reduced the cost of labor simultaneously, by original many people operation to now only need alone the operation can.
Drawings
FIG. 1 is a block diagram of the present invention in cooperation with a pipeline.
FIG. 2 is a structural diagram of the present invention, wherein the detection box is an internal structure.
FIG. 3 is a schematic view of the connection structure of the sliding rod and one of the camera mounting mechanisms according to the present invention.
Fig. 4 is a structural view of the camera mounting mechanism of the present invention.
Fig. 5 is a structural view of the slide board of the present invention.
Detailed Description
Refer to fig. 1 to 4. The utility model provides a TV backplate mesoscopic detection device, includes frame 1, and the first half of this frame 1 is equipped with detection case 2, and the lower half of this frame 1 is formed with the transmission space 3 that supplies assembly line 13 to pass. The detection box is characterized in that a sliding rod 4 is arranged in the transverse direction in the upper portion of the detection box 2, two camera mounting mechanisms 5 which are arranged at a left interval and a right interval are arranged on the sliding rod 4, the two camera mounting mechanisms 5 can be arranged on the sliding rod 4 in a sliding mode, the two camera mounting mechanisms 5 are both provided with a line scanning camera 6, the bottom of the detection box 2 is of a hollow structure, and the line scanning camera 6 can be arranged in a lifting mode. One side of the rack 1 is provided with an electric cabinet 7 with a control system inside, one side of the rack 1 is provided with a display screen 8, the output space 3 is internally provided with a photoelectric sensor 9 and an encoder 10, and the line scanning camera 6, the display screen 8, the photoelectric sensor 9 and the encoder 10 are all in signal connection with the control system.
Refer to fig. 3 to 5. Camera installation mechanism 5 includes riser 51, slider 52, slide 53 and camera mounting bracket 54, the rear end face of riser 51 is located to slider 52, slide 53 locates the preceding terminal surface of riser 51 and sets up with liftable, the preceding terminal surface of slide 53 is located to camera mounting bracket 54, the top and the camera mounting bracket 54 of line camera 6 are connected, slide bar 4 passes slider 52. In this embodiment, two slide rods 4 are provided, the two slide rods 4 are arranged at intervals up and down, two slide blocks 52 are provided for each camera mounting mechanism 5, and each slide block 52 corresponds to one slide rod 4. The camera mounting mechanism 5 can slide on the slide bar 4 from side to side as required by the cooperation of the slide block 52 and the slide bar 4, so as to adjust the left and right positions of the line scan camera 6.
Refer to fig. 3 to 5. The left side surface and the right side surface of the vertical plate 51 are respectively provided with a '>' shaped left sliding groove 511 and a '<' shaped right sliding groove (not marked in the figure), the rear end of the sliding plate 53 is provided with an accommodating groove 531, the left side surface of the accommodating groove 531 is a '/', the right side surface of the accommodating groove 531 is a '\\\' shape, the front end of the vertical plate 51 is matched with the accommodating groove 531, the accommodating groove 531 embraces the front end of the vertical plate 51, and the left end and the right end of the sliding plate 53 are respectively positioned in the left sliding groove 511 and the right sliding groove (not marked in the figure). The middle part of the vertical plate 51 is provided with a vertical screw rod groove 512, a screw rod 11 is arranged in the screw rod groove 512 in a penetrating manner, the bottom end of the screw rod 11 penetrates through the bottom of the vertical plate 51 to be connected with a hand wheel 12, the rear end face of the sliding plate 53 is provided with a screw rod sleeve (not marked in the figure), and the screw rod 11 penetrates through the screw rod sleeve (not marked in the figure). When the height of the line scan camera 6 needs to be adjusted, the hand wheel 12 is rotated to drive the screw rod 11 to rotate, and the screw rod 11 is matched with a screw rod sleeve (not shown in the figure) to drive the sliding plate 53 to lift, so as to drive the line scan camera 6 to lift.
A method for automatically detecting the inner appearance of a television backboard comprises the following steps:
1) the photoelectric sensor is used for sensing information of a television backboard on the water flow plate, when the control system receives shielding information of the water flow plate sensed by the photoelectric sensor, the two line-scan cameras are triggered to fetch images according to pulse frequency given by the encoder, when the water flow plate reaches the photoelectric sensor at the lower station and cannot sense the shielding information of the water flow plate, the image fetching of the two line-scan cameras is finished, the control system carries out picture arrangement processing on the images fetched by the two line-scan cameras, all the images fetched at this time are combined into one large image, and the large image is displayed on the display screen;
2) the control system creates a detection formula template of items to be detected of the television backboard, the detection formula template contains a type formula of each item to be detected, then six positioning points (three positioning points of each line scanning camera) are selected on a spliced large graph, the six positioning points are learned, the items to be detected are continuously learned after the learning of the positioning points is finished, the learning detection items are part of frames to be detected are selected, the corresponding type formula is selected, and the detection formula template is stored after all the detection items are learned;
3) and (3) repeating the step 1), starting to detect the television back plates of the same type, displaying the detection result on a display screen, respectively representing the correctness and the mistake of the detection result by using frame bodies with different colors, storing the correct picture into a designated folder after the detection is finished, storing the wrong picture into another folder, and counting all the items detected each time to facilitate the query.
The detection work in the step 3) comprises positioning and framing on the large image, products on the assembly line plate are manually placed, so that the positions of the products are relatively random, in order to solve the problem that the positions of the products cannot be unified, the positioning work is to adopt one line scanning camera to select three positioning points, the two line scanning cameras have six positioning points for affine transformation, a transformation matrix of the line scanning cameras is obtained through the positioning points, and then the position of each detection item is subjected to inverse matrix transformation to obtain the position of the corresponding detection item on the large image during detection; and the box selector is used for performing box selection on the detection item position on the large graph, when the detection item position is selected in a box mode, a corresponding type formula is selected, and after the type formula is selected, the control system automatically analyzes the detection result.
The detection items comprise black screws, small black screws, black screw black backgrounds, white screws, white screw black backgrounds, screw multi-lock, wire arranging adhesive tapes, light reflecting adhesive tapes, small adhesive tapes, adhesive tape damage and tilting, adhesive tape multi-paste, label offset, FFC, flexible flat cables, pins, small pins, buckles, piano covers, lamp holders, Boss columns, wire breakage, mirror surface white screws and detection of the existence of labels. The principle of all the detection items is as follows:
1. black screws:
and finding out black partial areas through a low threshold, screening out areas meeting conditions, calculating the areas of the areas, and comparing the obtained result values with the minimum and maximum areas of the algorithm parameters, wherein the result values are not wrong in the area range of the algorithm parameters, and otherwise, the result values are correct.
2. Black small screw
The algorithm is similar to the black screw, area regions meeting the conditions are found out through a threshold value, and then the area regions are compared with the area range of the algorithm parameters, if the area regions are not in the algorithm parameter range, the algorithm is wrong, and if the area regions are not in the algorithm parameter range, the algorithm is correct.
3. Black screw black background
Because the color of the detection item is very close to the background color, the contrast is firstly enhanced, then the area in the threshold value range is selected, and whether the area of the area is in the algorithm area parameter range or not is judged, the area is correct in the range, otherwise, the area is wrong.
4. White screw:
the white screw judgment method is to find a black hole, if the black hole meeting the conditions is found, the error is determined, otherwise, the error is determined. Therefore, if the misjudgment condition occurs, whether the observation result value is in the area range of the algorithm parameter or not is observed, and the range is adjusted in the range.
5. White screw black background
By utilizing the height difference between the screw edge and the back plate, under the effect of the light source, the screw edge has a circle of brighter area, the area is screened out and filled in the area of the calculation area, and the result value is correct in the algorithm parameter range, otherwise, the result value is wrong.
6. Multi-lock for screw
And searching a region (namely a black hole) meeting the condition through a low threshold, judging the area of the black hole region if the black hole is found, calculating the area, comparing the area with the area range of the algorithm parameter, wherein the area is correct within the area range, the error is determined when the area exceeds the maximum area, and the error is determined if the black hole is not found.
7. Wire-arranging adhesive tape
And screening out the regions according to the threshold range, wherein the region area result value is correct in the range, and otherwise, the region area result value is wrong.
8. Reflective adhesive tape
The method for expanding the threshold value is adopted because the light-reflecting adhesive tape has a light-reflecting phenomenon and the color of the light-reflecting part is close to that of the back plate, the light-reflecting adhesive tape is covered in the area within the threshold value range as far as possible, and after the area of the light-reflecting adhesive tape is obtained, the area is compared with the parameter area of the algorithm, the accuracy is achieved within the parameter range, and the error is avoided.
9. Small adhesive plaster
Because some rubberized fabric is much smaller than the normal rubberized fabric, the normal rubberized fabric algorithm can not cover the small rubberized fabric in area, so the small rubberized fabric algorithm is written additionally, the principle is similar to the two rubberized fabrics, the result value is correct in the parameter range of the area algorithm, otherwise, the result value is wrong.
10. Breakage and tilting of adhesive plaster
On the basis of the rubberized fabric, breakage and tilting detection are added, the tilting mainly detects whether four corners tilt, breakage is judged to be wrong if the area difference exceeds the value set by the algorithm parameters, and the breakage is judged to be correct if the area difference exceeds the value set by the algorithm parameters.
11. Adhesive plaster with multiple adhesive tapes
And adopting an adhesive tape algorithm to take an inverse method, judging that the adhesive tape is wrong, and otherwise, judging that the adhesive tape is correct.
12. Label multi-label
And (4) carrying out binarization processing on the picture of the label multi-labeling frame selection area, screening out the label area, and if the area of the area exceeds a certain value, judging that whether more labels are labeled wrongly, or else, judging that the labels are correct.
13. Label (offset)
The detection item needs to detect whether the label exists or not and whether the label deviates or not, when the template is manufactured and the label is selected in a frame mode, the position of the label is calculated through an algorithm, whether the label exists or not is judged through the algorithm during detection, the label is not directly wrong, the next step of calculating the position of the label is continued, if the position of the label exceeds a set range, the position of the label is wrong, and if the position of the label exceeds the set range, the position of the label is.
14、FFC
The detection method is an OCR (optical character recognition) method, when a formula is manufactured, after the FFC is selected in a frame, the characters on the FFC are recognized by an algorithm, and when the detection method is used, the characters are compared with a template, the same characters are correct, otherwise, the characters are wrong.
15. Flexible flat cable
This method is similar to the FFC detection method.
16、Pin
In order to prevent poor contact and product problems caused by the fact that the Pin flat cable is not inserted in place, the detection is increased, the Pin is white, therefore, when the area is selected, the range of a high threshold value is screened, after the area of the high threshold value is obtained, the horizontal straight line distance is calculated at one quarter and three quarters of the area, then the distance is compared with the template distance, the error is caused when the distance exceeds the range, and the error is caused otherwise.
17. Small Pin
And if the small Pin is the special Pin, the early-stage processing is required to be added, the region is converted into a region similar to the Pin, the later-stage algorithm is similar to the Pin, and whether the difference between the distance and the template distance exceeds the range is judged, the out-range is wrong, and otherwise, the out-range is correct.
18. Buckle
The buckle can also be called as special Pin, the processing method is similar to that of Pin, a buckle area is screened out through a high threshold value, the horizontal straight line distance between one quarter and three quarters of the buckle is calculated after image processing, the actual distance is compared with the template distance, the error is determined when the actual distance exceeds the template distance, and the error is determined otherwise.
19. Piano cover
The piano cover processing method is similar to Pin, the white areas of the cover and the flexible flat cable are screened out through a threshold value, then the white areas are negated in a picture to obtain the area between the cover and the flexible flat cable, the horizontal straight line distance between one quarter and three quarters of the area is calculated and compared with the template distance, the error is determined when the exceeding range is out, and the error is determined otherwise.
20. Lamp holder
The detection method adopts template matching, the selected position is selected as a template when a formula is manufactured, one template matching is carried out on the detected position when the formula is detected, the similarity reaches above the algorithm parameter score, and the result is correct, otherwise, the result is wrong.
21. Boss column
The Boss column adopts a method of matching circles, if the circles are matched, the Boss column is judged to be correct, otherwise, the Boss column is wrong.
22. Breakage of wire
By utilizing the copper exposure characteristic, a bright area exists after the part with copper exposure is imaged, and through a high threshold value and area screening method, if the area is satisfied, the copper exposure is judged, the result is wrong, otherwise, the result is correct.
23. Mirror face white screw
Because the mirror white screw has a light reflection phenomenon, a high threshold range is selected by utilizing the characteristic, whether the area of the area is larger than the minimum value of the area parameter of the algorithm is judged, the area larger than the minimum value of the area parameter of the algorithm is correct, and otherwise, the area is wrong.
24. Presence or absence of label
Because only whether the label exists or not is tested, the judgment is carried out by adopting the way that the label is pasted more and is reversed, the judgment that the label exists is correct, and the judgment that the label exists is wrong.
The above description is only an embodiment of the invention, but the design concept of the invention is not limited thereto, and any insubstantial modifications made by using the design concept should fall within the scope of infringing the invention.
Claims (7)
1. A television backboard appearance detection device is characterized in that: including the frame, the first half of this frame is equipped with the detection case, and the lower half of this frame is formed with the transmission space that supplies the assembly line to pass, be equipped with the slide bar on the horizontal direction in the upper portion of detection case, be equipped with the camera installation mechanism of two left and right interval arrangements on this slide bar, this two-phase machine installation mechanism can set up on the slide bar with sliding, two camera installation mechanism all are equipped with the line and sweep the camera, the bottom of detection case is hollow out construction, the line is swept the camera and is set up with liftable, one side of frame is equipped with the inside electric cabinet that has control system, and a side of this frame is equipped with the display screen, be equipped with photoelectric sensing ware and encoder in the output space, camera, display screen, photoelectric sensing ware and encoder are swept to the line all with control system signal connection.
2. The television backplane appearance inspection device of claim 1, wherein: camera installation mechanism includes riser, slider, slide and camera mounting bracket, the rear end face of riser is located to the slider, the preceding terminal surface and the liftable ground setting of riser are located to the slide, the preceding terminal surface of slide is located to the camera mounting bracket, the top and the camera mounting bracket connection of camera are swept to the line, the slider is passed to the slide bar.
3. The television backplane appearance inspection device of claim 2, wherein: the camera mounting mechanism is characterized in that the number of the sliding rods is two, the two sliding rods are arranged at intervals up and down, two sliding blocks are arranged on each camera mounting mechanism, and each sliding block corresponds to one sliding rod.
4. The television backplane appearance inspection device of claim 2, wherein: the left and right side faces of the vertical plate are respectively provided with a left sliding groove in a shape of '>' and a right sliding groove in a shape of '<', the rear end of the sliding plate is provided with an accommodating groove, the left side of the accommodating groove is in a shape of '/', the right side of the accommodating groove is in a shape of '\\' and is adaptive to the accommodating groove, the front end of the vertical plate is embraced by the accommodating groove, the left and right ends of the sliding plate are respectively positioned in the left sliding groove and the right sliding groove, the middle of the vertical plate is provided with a vertical screw rod groove, a screw rod is arranged in the screw rod groove in a penetrating mode, the bottom of the screw rod penetrates through the bottom of the vertical plate and is connected with a hand wheel, the rear end face of the sliding plate.
5. A method for automatically detecting the inner appearance of a television backboard is characterized by comprising the following steps: the television backplane appearance detection apparatus according to any one of claims 1 to 4, wherein the television backplane appearance automatic detection method comprises the following steps:
the photoelectric sensor is used for sensing information of a television backboard on the water flow plate, when the control system receives shielding information of the water flow plate sensed by the photoelectric sensor, the two line-scan cameras are triggered to fetch images according to pulse frequency given by the encoder, when the water flow plate reaches the photoelectric sensor at the lower station and cannot sense the shielding information of the water flow plate, the image fetching of the two line-scan cameras is finished, the control system carries out picture arrangement processing on the images fetched by the two line-scan cameras, all the images fetched at this time are combined into one large image, and the large image is displayed on the display screen;
the control system creates a detection formula template of items to be detected of the television backboard, the detection formula template contains a type formula of each item to be detected, then six positioning points are selected on a spliced big picture and are learned, after the learning of the positioning points is completed, the items to be detected are continuously learned, the learning of the detection items is to select partial frames to be detected, select corresponding type formulas, and after all the items to be detected are learned, the detection formula template is stored;
and (3) repeating the step 1), starting to detect the television back plates of the same type, displaying the detection result on a display screen, respectively representing the correctness and the mistake of the detection result by using frame bodies with different colors, storing the correct picture into a designated folder after the detection is finished, storing the wrong picture into another folder, and counting all the items detected each time to facilitate the query.
6. The method according to claim 5, wherein the method comprises the following steps: the detection work in the step 3) comprises positioning and framing on the large image, the positioning work is to adopt one line-scan camera to select three positioning points, the two line-scan cameras have six positioning points for affine transformation, a transformation matrix of the line-scan camera is obtained through the positioning points, and then the position of each detection item is transformed through an inverse matrix to obtain the position of the corresponding detection item on the large image during detection; and the box selector is used for performing box selection on the detection item position on the large graph, when the detection item position is selected in a box mode, a corresponding type formula is selected, and after the type formula is selected, the control system automatically analyzes the detection result.
7. The method according to claim 5, wherein the method comprises the following steps: the detection items comprise black screws, small black screws, black screw black backgrounds, white screws, white screw black backgrounds, screw multi-lock, wire arranging adhesive tapes, light reflecting adhesive tapes, small adhesive tapes, adhesive tape damage and tilting, adhesive tape multi-paste, label offset, FFC, flexible flat cables, pins, small pins, buckles, piano covers, lamp holders, Boss columns, wire breakage, mirror surface white screws and detection of the existence of labels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110113301.5A CN112730458A (en) | 2021-01-27 | 2021-01-27 | Automatic television backboard appearance detection device and detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110113301.5A CN112730458A (en) | 2021-01-27 | 2021-01-27 | Automatic television backboard appearance detection device and detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112730458A true CN112730458A (en) | 2021-04-30 |
Family
ID=75594207
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110113301.5A Pending CN112730458A (en) | 2021-01-27 | 2021-01-27 | Automatic television backboard appearance detection device and detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112730458A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1504742A (en) * | 2002-11-28 | 2004-06-16 | 威光机械工程股份有限公司 | Automatic optical detecting system for blemish assembly on printed circuit board |
| CN108120718A (en) * | 2016-11-29 | 2018-06-05 | 研祥智能科技股份有限公司 | Products detection system based on machine vision |
| CN108459020A (en) * | 2017-02-22 | 2018-08-28 | 惠州先进制造产业技术研究中心有限公司 | Multilayer surface backplate detection device |
| CN209879579U (en) * | 2019-06-14 | 2019-12-31 | 同方国际信息技术（苏州）有限公司 | Online vision inspection equipment of complete machine |
| CN210089626U (en) * | 2019-07-25 | 2020-02-18 | 苏州诠泰智能装备有限公司 | Automatic checking machine for assembly clearance of TV back plate |
| CN210834708U (en) * | 2019-10-28 | 2020-06-23 | 中设智能装备（苏州）有限公司 | Screen circuit detection equipment |
| CN210921997U (en) * | 2019-08-30 | 2020-07-03 | 上海奥塞尔材料科技有限公司 | Electric heating constant temperature drying cabinet of adjustable drying zone |
| CN111929239A (en) * | 2020-07-20 | 2020-11-13 | 浙江四点灵机器人股份有限公司 | AOI detection device and detection method for PCB part defects |
-
2021
- 2021-01-27 CN CN202110113301.5A patent/CN112730458A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1504742A (en) * | 2002-11-28 | 2004-06-16 | 威光机械工程股份有限公司 | Automatic optical detecting system for blemish assembly on printed circuit board |
| CN108120718A (en) * | 2016-11-29 | 2018-06-05 | 研祥智能科技股份有限公司 | Products detection system based on machine vision |
| CN108459020A (en) * | 2017-02-22 | 2018-08-28 | 惠州先进制造产业技术研究中心有限公司 | Multilayer surface backplate detection device |
| CN209879579U (en) * | 2019-06-14 | 2019-12-31 | 同方国际信息技术（苏州）有限公司 | Online vision inspection equipment of complete machine |
| CN210089626U (en) * | 2019-07-25 | 2020-02-18 | 苏州诠泰智能装备有限公司 | Automatic checking machine for assembly clearance of TV back plate |
| CN210921997U (en) * | 2019-08-30 | 2020-07-03 | 上海奥塞尔材料科技有限公司 | Electric heating constant temperature drying cabinet of adjustable drying zone |
| CN210834708U (en) * | 2019-10-28 | 2020-06-23 | 中设智能装备（苏州）有限公司 | Screen circuit detection equipment |
| CN111929239A (en) * | 2020-07-20 | 2020-11-13 | 浙江四点灵机器人股份有限公司 | AOI detection device and detection method for PCB part defects |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103048333B (en) | Appearance detection equipment and method | |
| CN108765416A (en) | PCB surface defect inspection method and device based on fast geometric alignment | |
| CN104749184A (en) | Automatic optical detection method and automatic optical detection system | |
| CN108982512A (en) | A kind of circuit board detecting system and method based on machine vision | |
| KR100910593B1 (en) | Faulty pixel detecting method and faulty pixel detecting system for led electric sign board | |
| JPH0727714A (en) | Inspection apparatus for plane display panel | |
| US6256066B1 (en) | High-resolution image pickup method and apparatus therefor | |
| CN106529365A (en) | Automatic pricing machine | |
| CN109887450B (en) | Artificial intelligent detection method for LED display screen | |
| CN112730458A (en) | Automatic television backboard appearance detection device and detection method | |
| CN111272776A (en) | Flexible circuit board defect detection method based on AI technology | |
| CN111474748A (en) | Machine vision inspection method and system for large-size L CD display module | |
| KR101426487B1 (en) | Apparatus for inspecting of display panel and method thereof | |
| CN217981299U (en) | Flexible circuit board visual detection device | |
| CN113758868A (en) | Automatic detection robot for display screen and detection method thereof | |
| CN206481360U (en) | A kind of large-scale workpiece scanner | |
| US20180365269A1 (en) | Method and apparatus for improving the efficiency of remote diagnosis of electronic and other equipment | |
| JP2969011B2 (en) | Appearance inspection device for soldering condition | |
| KR102286470B1 (en) | LED Digital Signage Defect Detection system through Similarity Analysis of Original Image Information and Displayed Image Information | |
| TWI502180B (en) | Brightness measuring method and system of device with backlight | |
| CN217111416U (en) | Lamp bead detection system | |
| CN220323138U (en) | Through type new forms of energy battery box CCD check out test set | |
| CN220230907U (en) | Display module assembly detects tool | |
| JPH0991428A (en) | Device for discriminating diameter of wire shaped material | |
| KR102599785B1 (en) | AI Vision Fault Detection System and Method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |