CN114496860A - IC photoetching character recognition and detection device surface imaging system and method - Google Patents

IC photoetching character recognition and detection device surface imaging system and method Download PDF

Info

Publication number
CN114496860A
CN114496860A CN202210131721.0A CN202210131721A CN114496860A CN 114496860 A CN114496860 A CN 114496860A CN 202210131721 A CN202210131721 A CN 202210131721A CN 114496860 A CN114496860 A CN 114496860A
Authority
CN
China
Prior art keywords
light source
chip
character
shaped
camera
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
Application number
CN202210131721.0A
Other languages
Chinese (zh)
Inventor
丁东红
黄荣
王凯
何宽芳
刘杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Foshan University
Original Assignee
Foshan University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Foshan University filed Critical Foshan University
Priority to CN202210131721.0A priority Critical patent/CN114496860A/en
Publication of CN114496860A publication Critical patent/CN114496860A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/67—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere
    • H01L21/67005—Apparatus not specifically provided for elsewhere
    • H01L21/67242—Apparatus for monitoring, sorting or marking
    • H01L21/67294—Apparatus for monitoring, sorting or marking using identification means, e.g. labels on substrates or labels on containers
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/67—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere
    • H01L21/67005—Apparatus not specifically provided for elsewhere
    • H01L21/67242—Apparatus for monitoring, sorting or marking
    • H01L21/67271—Sorting devices
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00—Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/67—Apparatus specially adapted for handling semiconductor or electric solid state devices during manufacture or treatment thereof; Apparatus specially adapted for handling wafers during manufacture or treatment of semiconductor or electric solid state devices or components ; Apparatus not specifically provided for elsewhere
    • H01L21/67005—Apparatus not specifically provided for elsewhere
    • H01L21/67242—Apparatus for monitoring, sorting or marking
    • H01L21/67282—Marking devices
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L22/00—Testing or measuring during manufacture or treatment; Reliability measurements, i.e. testing of parts without further processing to modify the parts as such; Structural arrangements therefor
    • H01L22/10—Measuring as part of the manufacturing process
    • H01L22/12—Measuring as part of the manufacturing process for structural parameters, e.g. thickness, line width, refractive index, temperature, warp, bond strength, defects, optical inspection, electrical measurement of structural dimensions, metallurgic measurement of diffusions
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L22/00—Testing or measuring during manufacture or treatment; Reliability measurements, i.e. testing of parts without further processing to modify the parts as such; Structural arrangements therefor
    • H01L22/20—Sequence of activities consisting of a plurality of measurements, corrections, marking or sorting steps

Landscapes

  • Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Computer Hardware Design (AREA)
  • Microelectronics & Electronic Packaging (AREA)
  • Power Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • General Physics & Mathematics (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses a surface imaging system and a method of an IC photoetching character recognition and detection device, which comprises a claw-shaped light source clamp, an arc-shaped light source clamp, a camera clamping mechanism, a portal frame fixing mechanism and a transmission mechanism, wherein the key of the imaging system comprises a bar-shaped light source, an annular light source, a backlight source, a telecentric lens and a CCD industrial camera, and the key of the related method comprises the positioning of the whole chip, the positioning of pins, the image data processing flow and the like. The IC photoetching character recognition and detection device surface imaging system and the method provide a method comprising a light source clamp, a camera clamp and a plurality of chip positioning recognition and realizing steps, wherein specific light source adjustment and camera angle adjustment need to be matched with a corresponding visual algorithm according to actual working conditions so as to realize the functions of chip surface photoetching character recognition which are more accurate, quicker and applicable under more conditions and further realize the functions of chip defect detection and classification.

Description

IC photoetching character recognition and detection device surface imaging system and method
Technical Field
The invention relates to the technical field related to semiconductor visual inspection, in particular to a surface imaging system and a surface imaging method for an IC photoetching character recognition and detection device.
Background
The chip is used in almost all electronic devices, and the character information on the chip is important information for marking the product. Therefore, the method is very important for distinguishing characters on a chip; therefore, the demand of the rapid character recognition technology applied to industrial production is particularly obvious; in particular to a visual design scheme of a visual detection device and an implemented algorithm flow. In the actual production process of the semiconductor chip, measurement of the pin size of the chip and the size of the whole chip and detection of a character mark on the surface of the chip are generally included; the detection of the character comprises the phenomena of recognizing the content of the character, the printing quality of the character, whether the character is missing or not and the like. Based on the requirement of character recognition detection, the character position needs to be positioned, and then further decision can be made; based on defect detection and size measurement, more effective image features of the chip need to be characterized.
Patent CN110097048A provides a method for fast image correction and character recognition of SOT chips, which can be used for fast recognition and detection of SOT chips.
Patent CN111274961A provides a character recognition and information analysis method for flexible IC substrate, which can be used for detecting variable-length character strings and recognizing adaptive environmental factors.
Patent CN113221889A provides a chip character anti-interference recognition method and device, which can effectively reduce partial interference of the environment when collecting images, prevent the chip from making an angle error, and improve the production efficiency.
The patent CN109685070A provides an image preprocessing method, which solves the technical problems in the prior art that most character detection in the IC sealing detection link is artificial inspection and has low accuracy.
Patent CN110852328A provides a method for recognizing characters on the back of a chip, which is characterized in that a lot of noise in a shot picture can be removed, and the chip does not need to be corrected.
Patent CN111330862A provides a dual positioning and adjusting device for chips, which can detect whether there is a problem on the upper and lower surfaces of the chip at the same time.
Patent CN111754461A provides a method and device for positioning image character area of semiconductor chip; by providing a corner filtering method and improving a convex hull detection algorithm, the image character area of the semiconductor chip can be more accurately positioned.
Patent CN107767373A provides a method for visual detection of fast sub-pixel precision of chip pins, which realizes fast high-precision detection of chip pin width.
The recognition of the characters on the surface of the chip inevitably has some defects of the characters, such as the loss of the characters, the unobvious character printing, the influence of scratches on the surface of the chip on the character recognition and the like. Some of the defects can reduce the difficulty of visual algorithm design through light path design, and further can easily classify the defect types and make corresponding visual algorithm decisions. Meanwhile, each part on the chip has characteristics which can be characterized under different conditions, so that a light source with a specific angle needs to be designed for illumination, and the specific characteristic information can be characterized. Illumination of locations such as chip pins, chip characters, chip sides, chip overall surfaces, etc.; the illumination at different positions needs to consider that the combination of the whole illumination system is not influenced mutually, and meanwhile, the design of a visual algorithm is matched, so that the applicability of different chips under different environments can be adapted.
Patent CN111366588A provides a semiconductor chip character surface defect pin defect detection light source, has solved the three light sources of old version and has superimposed the problem in the same track, compatible other product detection simultaneously.
Patent CN213516936U provides a detection mechanism of an IC chip detection device, which can adjust the corresponding mechanism to realize the illumination of the light source to the character or the pin, so as to achieve a better detection effect.
Patent CN210629661U provides an OCR character collection system based on degree of depth example topic vision, and wherein the illumination angle of camera, light source and the locating place of character carrier all can quick adjustment, realize the effect of quick and accurate collection character.
Patent CN104580860A provides an auxiliary device for reading characters on the surface of a chip, which realizes the preprocessing of chip image acquisition, and makes the chip acquire images quickly and stably.
Patent CN104881665A provides a method and apparatus for recognizing and verifying chip characters, which reduces the influence of the problems of character fracture and adhesion on character recognition.
Patent CN110927184A provides a visual inspection method for chip packaging defects, which achieves the effects of good imaging, accurate positioning and high inspection efficiency.
Patent CN108982508A provides a method for detecting defects of a plastic package IC chip based on feature template matching and deep learning, which solves the problems of chip positioning and character detection, and detects and classifies defects of chip pins, and meets certain requirements for online detection of IC chips.
Based on the descriptions of the above various patents, a number of targeted chip identification and defect classification problems are essentially solved. However, the function implementation of each patent has certain applicability to a particular type of chip. The detection and identification device of a semiconductor chip system and the simple algorithm design application of the system, in particular to a detection system which is matched with the operation of a plurality of cameras, the light path design of a plurality of light sources and a vision algorithm, and no chip detection vision system which can be specifically suitable for various environments exists in the market.
Aiming at the requirements of character recognition and defect detection of a chip, characterization and measurement of chip pin characteristics and the like; the application provides an IC photoetching character recognition and detection device, a surface imaging system and a surface imaging method.
Disclosure of Invention
The invention aims to provide a surface imaging system and a surface imaging method for an IC photoetching character recognition and detection device, which aim to solve the problems that a type of chip needs to be detected on a plurality of devices, the efficiency is low when the character of the chip is judged to be correct manually, and the traditional light source clamp and the traditional camera clamp lack multi-direction angle adjustment.
In order to achieve the purpose, the invention provides the following technical scheme: a surface imaging system of an IC photoetching character recognition and detection device comprises a claw-shaped light source clamp which is vertically arranged, an arc-shaped light source clamp which is arranged on the side surface, a camera clamping mechanism with multi-directional adjustment, a gantry supporting frame and a transmission mechanism;
the surface imaging system comprises a strip light source, a low-angle annular light source, a backlight source, a telecentric lens and a CCD industrial camera;
the related light source controls the illumination of the light source according to the characteristics represented by the requirements, namely whether the light source illuminates the IC sample and the illumination brightness can be controlled by coding different characteristics, so that the waste of the light source and certain influence possibly caused are avoided.
Preferably, the head end of the vertically arranged claw-shaped light source clamp is connected with the gantry support frame, a left-right rotation adjusting structure is formed between the whole vertically arranged claw-shaped light source clamp and the gantry support frame through a rotation fixing block, the rotation fixing block is connected with a claw-shaped light source fixing frame through a connecting rod structure of a U-shaped connecting block, the head end of the U-shaped connecting block is connected with a convex connecting block integrated at the side end of the rotation fixing block, the vertically arranged claw-shaped light source clamp can form an adjusting structure of a pitching angle through the convex connecting block, the tail end of the U-shaped connecting block is connected with the claw-shaped light source fixing frame, the claw-shaped light source fixing frame forms a further telescopic action adjusting structure, and hole structures are uniformly distributed on the claw-shaped light source fixing frame and are used for mounting various bar-shaped light sources and annular light sources, and the strip light source and the ring light source are used for illumination of the IC sample.
By adopting the technical scheme, all directions of the whole mechanism are provided with the positions for fixing the light sources, the whole mechanism can be connected with the corresponding light sources through pins, and the whole mechanism has a certain space position adjusting function. Through the adjusting function of the light source clamp, the light source can reach a more ideal illumination angle in the space. The light source clamp adjustment mainly comprises two major steps, wherein one step is primary adjustment, and the light source clamp has the functions of left-right, pitching and stretching adjustment; and the second is precise adjustment, namely further fine adjustment between the light source and the claw-shaped fixing frame through pin connection.
Preferably, the arc-shaped light source clamp main body arranged on the side face is an arc-shaped honeycomb plate, the arc-shaped honeycomb plate is integrally in an arc shape, and honeycomb-shaped fixing holes are formed in the arc-shaped honeycomb plate and used for fixing corresponding backlight light sources; the arc-shaped honeycomb plate is connected with the gantry support frame, and the arc-shaped honeycomb plate forms a pitching angle adjusting structure; the backlight light source is connected with the holes on the honeycomb plate through the U-shaped connecting rod to form an angle adjusting structure; the arc-shaped cellular boards are symmetrically arranged on two sides of the conveying mechanism to achieve the symmetrical lighting effect on the IC samples.
By adopting the technical scheme, the arc-shaped light source clamp placed on the side surface is integrally in an arc shape, wherein the arc plate is provided with the honeycomb-shaped fixing holes so as to fix the corresponding light source. The arc-shaped plate is connected with the portal frame, and the pitching angle can be adjusted; furthermore, the light source is connected with the holes on the honeycomb plate through pins, and fine angle adjustment is carried out. The adjusting step of the arc-shaped light source clamp placed on the side surface is mainly divided into two steps, wherein the first step is pitching adjustment; and the second step is fine adjustment of the honeycomb plate connected with the light source and the pin.
Preferably, the CCD industrial camera is fixedly arranged at the tail end of the camera clamping mechanism with multi-direction adjustment, the mechanism at the tail end of the camera clamping mechanism with multi-direction adjustment mainly comprises three rectangular honeycomb fixing plates, the rectangular honeycomb fixing plates mainly play a role in fixing the position of the CCD industrial camera, a plurality of cameras can be fixed by the rectangular honeycomb fixing plates, the CCD industrial camera and the rectangular honeycomb fixing plates are connected through camera fixing connecting blocks, the camera fixing connecting blocks form a plane rotation angle adjusting structure, the rectangular honeycomb fixing plates form a rotating structure around a middle camera fixing connecting block to realize spatial multi-direction angle adjustment of the camera, the head end of each rectangular honeycomb fixing plate is connected with a rack connecting guide block, and the camera clamping mechanism with multi-direction adjustment forms a pitching angle adjusting structure at the front end of the rack connecting guide block, and rack structure is connected to rack connection guide block end fixedly, rack structure that rack connection guide block end is fixed is the guide rail structure that the rectangle protection shield is connected, and rack structure that rack connection guide block end is fixed passes through rack and pinion adjust knob regulation rack and pinion's stroke control camera anchor clamps terminal seesaw, camera clamping mechanism whole that has diversified regulation constitutes left and right control structure in gantry support frame department.
By adopting the technical scheme, the tail end of the camera clamping mechanism with multi-direction adjustment mainly comprises three honeycomb-shaped fixing plates, and the honeycomb holes are mainly used for fixing the position of the camera. The gear rack knob can be further adjusted and controlled to control the front and back movement of the tail end of the camera clamp. The camera fixture adjustment mainly comprises two major steps, wherein one step is primary adjustment, and the camera fixture has the functions of left-right adjustment, pitching adjustment and front-back adjustment; and the second is precise adjustment, namely further fine adjustment between the CCD industrial camera and the honeycomb fixing plate through a pin.
A IC photoetching character recognition and detection device surface imaging method comprises the steps of positioning the whole chip, positioning pins and processing image data;
the positioning method of the whole chip mainly comprises the steps of highlighting the whole characteristics of the chip and removing unnecessary noise by matching illumination imaging of a light source, extracting a bright area by adopting a binarization algorithm, further calculating geometric information of the bright area, and performing affine transformation to correct the placement position of the chip. Further, the light source is selected from a low-angle annular light source, a bar light source and a backlight light source;
the pin positioning method scans the chip pins in a form of forming scanning lines so as to judge the specific placing positions of the chip pins. Furthermore, a positioning method combining template matching and feature analysis is provided, so that the angle required to be converted to the correct position can be further corrected, and the chip positioning and identifying function can be realized;
the image data processing flow method comprises the steps that the image preprocessing in the previous stage mainly comprises the transformation of the image on a color space, so that the average data of all the images is obtained; and judging the pre-processing image effect graph selected for decision by setting a threshold range.
Preferably, in the IC lithography character recognition process, in order to accurately and quickly recognize characters, a certain requirement is imposed on the recognized character pose, so that an algorithm is further designed to correct the character pose. In the positive rectangular image, characters in some rectangular images may be in an inverted posture, and a positioning algorithm is further designed to ensure that the characters are in an upright state.
The chip-based feature analysis and calculation method has two methods: firstly, whether the current condition of the chip is the upright posture or not is judged by screening and counting the number of pins of the chip. Secondly, forming a horizontal scanning line on the image of the rectangular area, scanning from top to bottom, and intersecting with the pin characteristics; and judging whether the current condition of the chip is in the upright posture or not according to the intersection result. The initial position formed by the scanning lines and the working interval of the scanning lines are set, which are the keys for ensuring accurate scanning and improving the running speed; the specific threshold value needs to be calculated with reference to the actual chip position.
Preferably, in the method for positioning the pins by using the horizontal scanning lines, the pins of the chip are scanned mainly in a form of forming the scanning lines, so as to judge the specific placement positions of the pins of the chip. Further, the initial position of the scan line formation is to set an initial coordinate threshold by calculating the center position of the characters on the chip surface, and further to set a spacing threshold between the scan lines. The main role of threshold setting is as follows: firstly, the threshold setting can avoid the formation of unnecessary scanning lines, thereby saving the consumption of a computer memory and further saving the consumption of time; and secondly, due to the interval arrangement among the scanning lines, the misjudgment caused by different lengths of the chip pins can be effectively avoided, meanwhile, the time is saved for the computer scanning, and the operation speed of the algorithm is further improved. Thirdly, in the scanning process, through the setting of threshold value, can judge the concrete position of chip pin, further can realize the function of pin measurement, further can get the front and back contrast of the result of intersection according to scan line and pin, further can judge whether the chip pin has the condition of lacking the pin, realizes the defect detection function of chip pin.
Preferably, the defect detection function of the chip pins is to mark the corresponding result correspondingly when performing corresponding image preprocessing, so as to realize the effect of defect classification. Such as:
first, when selecting the best processing channel, the distribution of the brightest pixels of the character is calculated, and the result after calculation can be used to judge whether the character surface is printed fully or lightly. If the printing is full, the value of the brightest pixel obtained by calculation is higher, otherwise, the obtained result value is lower.
Secondly, when the connected domains of the number of the characters are counted, whether the characters printed by the current chip are lost or not can be judged according to the current counting result.
Before obtaining the whole connected domain of the character, the whole character needs to be subjected to difference processing, the standard character and the character which needs to be identified currently are subjected to difference processing, and the difference result can be used for judging whether the current character is in the situations of few strokes, multiple strokes or breakage and the like.
In the statistical classification of defect classification, all the defect classification is carried out by making corresponding algorithm marks in necessary image preprocessing for decision classification, and large-scale neural network training is not required, so that a better and faster classification and identification effect is achieved.
Compared with the prior art, the invention has the beneficial effects that: the IC photoetching character recognition and detection device, the surface imaging system and the method,
1. by adjusting the light source clamp and the CCD industrial camera clamp, the light source and the camera can present various lighting and shooting postures in space, the operation under various complex environments of actual industrial operation can be met, various implementation methods and flow steps are provided, and a chip detection visual system capable of being specifically suitable for various environments is provided;
2. the problems that one type of chip needs to be detected on multiple devices, the efficiency is low when the chip characters are correct or not is judged manually, and the traditional light source clamp and the traditional camera clamp lack multi-direction angle adjustment are solved.
Drawings
FIG. 1 is a schematic view of a vertically disposed "claw" type light source fixture according to the present invention;
FIG. 2 is a schematic view of a laterally disposed curved light source fixture according to the present invention;
FIG. 3 is a schematic view of a camera clamping mechanism with multi-directional adjustment according to the present invention;
FIG. 4 is a general flowchart of the IC character recognition method of the present invention;
FIG. 5 is a schematic diagram of a template matching and edge angle adjustment process according to the present invention;
FIG. 6 is a schematic view of the annular light source illumination of the present invention;
FIG. 7 is a flow chart of the character normalization process of the present invention;
FIG. 8 is a schematic diagram of a defective character classification system according to the present invention;
FIG. 9 is a scan line algorithm flow chart of the present invention;
FIG. 10 is a flow chart of a feature analysis location identification algorithm of the present invention;
fig. 11 is a schematic view of the overall structure of the present invention.
In the figure: 1. a vertically disposed "claw" type light source fixture; 101. rotating the fixed block; 102. a convex connecting block; 103. a U-shaped connecting block; 104. a claw-shaped light source fixing frame; 2. the arc-shaped light source clamp is arranged on the side surface; 201. a U-shaped connecting rod; 202. an arc-shaped honeycomb plate; 3. the camera clamping mechanism is provided with multi-direction adjustment; 301. a rack and pinion adjustment knob; 302. the rack is connected with the guide block; 303. a camera fixed connection block; 304. a rectangular honeycomb fixing plate; 4. a gantry support frame; 5. a transport mechanism; 6. a strip light source; 7. an annular light source; 8. a backlight light source; 9. a telecentric lens; 10. a CCD industrial camera; 11. IC samples.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-11, the present invention provides a technical solution: a IC photoengraving character discerns and the imaging system of the surface of checkout gear, IC photoengraving character discerns and checkout gear includes the "claw" type light source clamp 1 placed vertically, the arcuate light source clamp 2 placed laterally, have camera clamping mechanism 3 that the multi-azimuth adjusts, the gantry support frame 4, the transport mechanism 5;
the surface imaging system comprises a bar-shaped light source 6, a low-angle annular light source 7, a backlight light source 8, a telecentric lens 9 and a CCD industrial camera 10;
the lighting of the light source is controlled by the characteristics represented by the requirements, that is, whether the light source illuminates the IC sample 11 and the brightness of the lighting can be controlled by encoding different characteristics, so that the waste of the light source and possible influence can be avoided.
The head end of a vertically placed claw-shaped light source clamp 1 is connected with a gantry support frame 4, a left-right rotation adjusting structure is formed between the whole vertically placed claw-shaped light source clamp 1 and the gantry support frame 4 through a rotation fixing block 101, the rotation fixing block 101 is connected with a claw-shaped light source fixing frame 104 through a connecting rod structure of a U-shaped connecting block 103, the head end of the U-shaped connecting block 103 is connected with a convex connecting block 102 integrated at the side end of the rotation fixing block 101, the vertically placed claw-shaped light source clamp 1 can form a pitching angle adjusting structure through the convex connecting block 102, the tail end of the U-shaped connecting block 103 is connected with a claw-shaped light source fixing frame 104, the claw-shaped light source fixing frame 104 forms a further telescopic action adjusting structure, hole structures are uniformly distributed on the claw-shaped light source fixing frame 104 and are used for installing various bar-shaped light sources 6 and annular light sources 7, and a bar light source 6 and a ring light source 7 are used for illumination of an IC sample 11, a schematic view of a vertically placed "claw" type light source holder 1 is shown in fig. 1.
The whole mechanism has the advantages that the positions of the light sources are fixed in all directions, the corresponding light sources can be connected through pins, and the whole mechanism has a certain space position adjusting function. Through the adjusting function of the light source clamp, the light source can reach a more ideal illumination angle in the space. The light source clamp adjustment mainly comprises two major steps, wherein one step is primary adjustment, and the light source clamp has the functions of left-right, pitching and stretching adjustment; and the second is precise adjustment, and the light source and the claw-shaped fixing frame are further finely adjusted through pin connection.
The main body of the arc-shaped light source clamp 2 arranged on the side face is an arc-shaped cellular board 202, the arc-shaped cellular board 202 is integrally in an arc shape, and the arc-shaped cellular board 202 is provided with honeycomb-shaped fixing holes for fixing corresponding backlight light sources 8; the arc cellular board 202 is connected with the gantry support frame 4, and the arc cellular board 202 forms a pitching angle adjusting structure; the backlight light source 8 is connected with the holes on the honeycomb panel through a U-shaped connecting rod 201 to form an angle adjusting structure; the arc-shaped honeycomb plates 202 are symmetrically arranged at two sides of the conveying mechanism 5 to achieve a symmetrical illumination effect on the IC sample 11, and a schematic diagram of the arc-shaped light source clamp 2 arranged at the side is shown in FIG. 2.
The laterally placed arc-shaped light source clamp 2 is integrally in an arc shape, wherein the arc plate is provided with honeycomb-shaped fixing holes so as to fix the corresponding light source. The arc-shaped plate is connected with the portal frame, and the pitching angle can be adjusted; furthermore, the light source is connected with the holes on the honeycomb plate through pins, and fine angle adjustment is carried out. The adjusting step of the arc-shaped light source clamp placed on the side surface is mainly divided into two steps, wherein the first step is pitching adjustment; and the second step is fine adjustment of the honeycomb plate connected with the light source and the pin.
The CCD industrial camera 10 is fixedly arranged at the tail end of the camera clamping mechanism 3 with multi-direction adjustment, the mechanism at the tail end of the camera clamping mechanism 3 with multi-direction adjustment mainly comprises three rectangular honeycomb fixing plates 304, the rectangular honeycomb fixing plates 304 mainly play a role in fixing the position of the CCD industrial camera 10, a plurality of cameras can be fixed by the rectangular honeycomb fixing plates 304, the CCD industrial camera 10 is connected with the rectangular honeycomb fixing plates 304 through camera fixing connecting blocks 303, the camera fixing connecting blocks 303 form a plane rotation angle adjusting structure, the rectangular honeycomb fixing plates (304) form a rotating structure around the middle camera fixing connecting blocks 303 to realize the spatial multi-direction angle adjustment of the cameras, the head end of each rectangular honeycomb fixing plate (304) is connected with a rack connecting guide block 302, and the camera clamping mechanism 3 with multi-direction adjustment forms a pitching angle adjusting structure at the front ends of the rack connecting guide blocks 302, and rack connecting guide block 302 end fixedly connected with rack structure, rack structure that rack connecting guide block 302 end is fixed is the guide rail structure that the rectangle protection shield connects, and rack structure that rack connecting guide block 302 end is fixed adjusts the stroke control camera anchor clamps end back and forth movement through rack and pinion adjust knob 301, camera clamping mechanism 3 that has diversified regulation wholly constitutes left and right control structure in longmen support frame 4 department, camera clamping mechanism 3 that has diversified regulation is the sketch map as shown in fig. 3.
The tail end of the camera clamping mechanism 3 with multi-direction adjustment mainly comprises three honeycomb-shaped fixing plates, and the honeycomb holes are mainly used for fixing the position of the camera. The gear rack knob can be further adjusted and controlled to control the front and back movement of the tail end of the camera clamp. The camera fixture adjustment mainly comprises two major steps, wherein one step is primary adjustment, and the camera fixture has the functions of left-right adjustment, pitching adjustment and front-back adjustment; and the second is precise adjustment, and the CCD industrial camera and the honeycomb fixing plate are further finely adjusted through a pin.
A IC photoetching character recognition and detection device surface imaging method comprises the steps of positioning the whole chip, positioning pins and processing image data;
the positioning method of the whole chip mainly comprises the steps of highlighting the whole characteristics of the chip and removing unnecessary noise by matching illumination imaging of a light source, extracting a bright area by adopting a binarization algorithm, further calculating geometric information of the bright area, and performing affine transformation to correct the placement position of the chip. Further, the light sources are selected from a bar light source 6, a low-angle annular light source 7 and a backlight light source 8;
the pin positioning method scans the chip pins in a form of forming scanning lines so as to judge the specific placing positions of the chip pins. Furthermore, a positioning method combining template matching and feature analysis is provided, so that the angle required to be converted to the correct position can be further corrected, and the chip positioning and identifying function can be realized;
the image data processing flow method includes the steps of performing color space transformation on an image to obtain average data of all the images; and judging the pre-processing image effect graph selected for decision by setting a threshold range.
In the image data processing flow method, the image acquired by the imaging system does not necessarily completely conform to the design of the visual algorithm, so the quality of the acquired image needs to be further optimized. The former image preprocessing mainly includes transforming the image in color space to obtain the average data of all the images; setting a threshold range, and judging the effect graph of the pre-processing image to be selected.
And splitting an image obtained from the camera into an R, G and B three-channel image, and further converting the image into an H, S and V three-channel image. And calculating all gray values at each coordinate point by using the 6 images generated by calculation, wherein the result is an average value of the gray values, setting a threshold value, judging and determining an optimal processing channel, and finally outputting a single-channel image. The specific calculation formula of the average gray value is described as follows, and let the gray value of each channel image be respectively expressed as fR(x，y)、fG(x，y)、fB(x，y)、fH(x，y)、fS(x，y)、fV(x, y) averaging the gray values of the channels, and finally adding the average values of the channels to obtain the gray value of the image of the single channel, namely the gray value of the image of the single channel
Figure BDA0003502917480000121
In the formula, n represents the number of channels of the current input image, and ux and y are output images. When the formula is actually applied, the formula on the molecule is not necessarily the listed input image, but the input of the corresponding channel is properly removed or added according to the actual situation; but in general, the mode of the equation is applied for the calculation.
Further, if the average gray-level value calculated in the formula does not reach the set threshold range, the image channel is reselected and the above steps are repeated until the desired optimal processing channel is output.
The determined optimal image processing channel can be further calculated to form a minimum circumscribed rectangle with a direction; and calculating the angle and the position of the minimum circumscribed rectangle, and translating and rotating the minimum circumscribed rectangle to obtain a rectangular region of the character ROI region. So far, the initial positioning already obtains the position of the whole chip.
In order to further correct the arrangement position of the chip, a shape-based template matching algorithm is also proposed to further determine the positions and angles of the character features. Since extreme positions occur when calculating the angle and position coordinates for salient features, this can lead to errors in position correction and thus further to recognition errors. If the placement position of the IC is exactly at the calculated critical angle position, there may be a position where the correct character placement position cannot be further obtained, but is different from the expected effect by 90 °. The critical position of (c) may occur when the IC top and bottom pin centerlines are at an angle of 80 deg. -90 deg. to the vertical centerline or at an angle of 0 deg. -10 deg. to the horizontal centerline. The specific angle correction method and steps are as follows: firstly, obtaining a preliminarily positioned rectangular image, then calculating a sideline angle theta 1 of a rectangular area of a character in the middle of a chip, secondly, obtaining a template angle theta 2 found by template matching, obtaining an angle theta by mutual difference of the obtained angles, and further judging the alignment state of the current chip according to the angle theta. If the value of θ is not within the predetermined range of thresholds, the above operations are repeated until the desired value is obtained. A specific implementation flowchart is shown in fig. 5.
In a visual imaging system module, the illumination of the light source is critical to highlighting the target feature. The selection of the color of the light source, the angle and distance of the light source illumination, etc. may cause certain imaging difficulties for the imaging system. The preferred light source is a white annular light source, the irradiation angle of a lamp bead and a sample form an angle of 0-15 degrees, and the distance between the light source and the lens is-150 mm-250 mm. The distance is defined by using the center point of the object as the mass point. If the lens takes the intersection point of the horizontal center and the vertical center as a mass point, the light source is similar. The distance between the two is described, a coordinate system is established by taking the two exactly coincident ends as a defining line, the upward movement is positive, and the downward movement is negative. The angles involved are developed and explained with the horizontal plane being 0 °. The specific light source implementation is schematically shown in fig. 6 below.
In the character feature segmentation process, a certain adhesion or fracture condition may exist in the connected domain of the character. The characters need to be normalized to achieve the integrity and unity of the characters. The realization method and the steps are as follows: and forming a minimum circumscribed rectangle for the obtained character-containing area, setting a segmentation threshold according to the width and height of the character to obtain an area in which each character is relatively independent, further intersecting the character with the corresponding pixel position of the character-containing area of the original drawing, and finally obtaining the completely segmented character area. The specific algorithm flow is shown in fig. 7 below.
In the IC photoetching character recognition process, in order to accurately and quickly recognize characters, certain requirements are required on the recognized character postures, so that an algorithm is further designed to correct the character postures. In the positive rectangular image, characters in some rectangular images may be in an inverted posture, and a positioning algorithm is further designed to ensure that the characters are in an upright state.
The chip-based feature analysis and calculation method has two methods: firstly, whether the current condition of the chip is the upright posture or not is judged by screening and counting the number of pins of the chip. Secondly, forming a horizontal scanning line on the image of the rectangular area, scanning from top to bottom, and intersecting with the pin characteristics; and judging whether the current condition of the chip is in the upright posture or not according to the intersection result. The initial position formed by the scanning lines and the working interval of the scanning lines are set, which are the keys for ensuring accurate scanning and improving the running speed; the specific threshold needs to be calculated with reference to the actual chip position, and the specific algorithm flows of the two positioning methods are shown in fig. 10 below.
In the method for positioning the pins by the horizontal scanning lines, the pins of the chip are scanned mainly in a form of forming the scanning lines, so that the specific placing positions of the pins of the chip are judged. Further, the initial position of the scan line formation is to set an initial coordinate threshold by calculating the center position of the characters on the chip surface, and further to set a spacing threshold between the scan lines. The main role of threshold setting is as follows: firstly, the threshold setting can avoid the formation of unnecessary scanning lines, thereby saving the consumption of a computer memory and further saving the consumption of time; and secondly, due to the interval arrangement among the scanning lines, the misjudgment caused by different lengths of the chip pins can be effectively avoided, meanwhile, the time is saved for the computer scanning, and the operation speed of the algorithm is further improved. Thirdly, in the scanning process, the specific position of the chip pin can be judged through the setting of the threshold value, the function of pin measurement can be further realized, the comparison before and after the intersection result of the scanning line and the pin can be further obtained, the condition whether the chip pin has the pin shortage can be further judged, the defect detection function of the chip pin can be realized, and the specific scanning algorithm flow is as shown in the following fig. 9.
The chip pin defect detection function is to mark the corresponding result correspondingly when performing the corresponding image preprocessing, so as to realize the effect of defect classification. Such as:
first, when selecting the optimal processing channel, the distribution of the brightest pixels of the character is calculated, and the result after calculation can be used to determine whether the character surface is printed fully or lightly. If the printing is full, the value of the brightest pixel is calculated to be higher, otherwise, the obtained result value is lower.
Secondly, when the connected domains of the number of the characters are counted, whether the characters printed by the current chip are lost or not can be judged according to the current counting result.
Before obtaining the whole connected domain of the character, the whole character needs to be subjected to difference processing, the standard character and the character which needs to be identified currently are subjected to difference processing, and the difference result can be used for judging whether the current character is in the situations of few strokes, multiple strokes or breakage and the like.
In the statistical classification of defect classification, all the classification is carried out by making corresponding algorithm marks in the necessary image preprocessing and large-scale neural network training is not needed, so that better and faster classification and identification effects are achieved, and specific defect chip classification diagrams are shown in fig. 8.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.

Claims (8)

1. A surface imaging system of an IC photoetching character recognition and detection device is characterized in that: the IC photoetching character recognition and detection device comprises a claw-shaped light source clamp (1) which is vertically arranged, an arc-shaped light source clamp (2) which is laterally arranged, a camera clamping mechanism (3) with multi-directional adjustment, a gantry support frame (4) and a transmission mechanism (5);
the surface imaging system comprises a bar-shaped light source (6), a low-angle annular light source (7), a backlight light source (8), a telecentric lens (9) and a CCD industrial camera (10);
the related light sources control the illumination of the light sources according to the characteristics characterized by the requirements, namely whether the light sources illuminate the IC sample (11) or not and the illumination brightness can be controlled by coding different characteristics to be characterized, so that the waste of the light sources and certain influence possibly caused are avoided.
2. The IC lithographic character recognition and detection apparatus surface imaging system according to claim 1, wherein: the head end of the vertically placed claw-shaped light source clamp (1) is connected with the gantry support frame (4), and a left-right rotation adjusting structure is formed between the whole vertically placed claw-shaped light source clamp (1) and the gantry support frame (4) through a rotation fixing block (101);
the rotary fixing block (101) is connected with a claw-shaped light source fixing frame (104) through a connecting rod structure of a U-shaped connecting block (103), the head end of the U-shaped connecting block (103) is connected with a convex connecting block (102) integrated with the side end of the rotary fixing block (101), and a vertically-arranged claw-shaped light source clamp (1) can form a pitching angle adjusting structure through the convex connecting block (102);
the tail end of the U-shaped connecting block (103) is connected with a claw-shaped light source fixing frame (104), the claw-shaped light source fixing frame (104) forms a further telescopic action adjusting structure, hole structures are uniformly distributed on the claw-shaped light source fixing frame (104) and used for mounting various strip-shaped light sources (6) and annular light sources (7), and the strip-shaped light sources (6) and the annular light sources (7) are used for illuminating an IC sample (11).
3. The IC lithographic character recognition and detection apparatus surface imaging system according to claim 1, wherein: the main body of the arc-shaped light source clamp (2) arranged on the side face is an arc-shaped cellular board (202), the arc-shaped cellular board (202) is integrally in an arc shape, and the arc-shaped cellular board (202) is provided with honeycomb-shaped fixing holes for fixing corresponding backlight light sources (8);
the arc-shaped cellular board (202) is connected with the gantry support frame (4), and the arc-shaped cellular board (202) forms a pitching angle adjusting structure;
the backlight light source (8) is connected with the holes on the cellular board through a U-shaped connecting rod (201) to form an angle adjusting structure;
the arc-shaped cellular boards (202) are symmetrically arranged at two sides of the conveying mechanism (5) to achieve a symmetrical illumination effect on the IC sample (11).
4. The IC lithographic character recognition and detection apparatus surface imaging system according to claim 1, wherein: the CCD industrial camera (10) is fixedly mounted at the tail end of the camera clamping mechanism (3) with multi-direction adjustment, the mechanism at the tail end of the camera clamping mechanism (3) with multi-direction adjustment mainly comprises three rectangular honeycomb fixing plates (304), the rectangular honeycomb fixing plates (304) mainly play a role in fixing the position of the CCD industrial camera (10), the rectangular honeycomb fixing plates (304) can fix a plurality of cameras, the CCD industrial camera (10) is connected with the rectangular honeycomb fixing plates (304) through camera fixing connecting blocks (303), and the camera fixing connecting blocks (303) form a plane rotation angle adjusting structure;
the rectangular honeycomb fixing plate (304) forms a rotating structure around the middle camera fixing connecting block (303) so as to realize spatial multi-directional angle adjustment of the camera;
the head end of the rectangular honeycomb fixing plate (304) is connected with a rack connecting guide block (302), a camera clamping mechanism (3) with multi-direction adjustment is arranged at the front end of the rack connecting guide block (302) to form a pitching angle adjusting structure, and the tail end of the rack connecting guide block (302) is fixedly connected with a rack structure;
the rack structure fixed at the tail end of the rack connecting guide block (302) is a guide rail structure connected with a rectangular protection plate, the rack structure fixed at the tail end of the rack connecting guide block (302) adjusts the stroke of a gear rack through a gear rack adjusting knob (301) to control the front and back movement of the tail end of a camera clamp, and the camera clamping mechanism (3) with multi-directional adjustment integrally forms a left and right adjusting structure at the gantry support frame (4).
5. A surface imaging method of an IC photoetching character recognition and detection device is characterized in that: the imaging method comprises the steps of positioning the whole chip, positioning pins and processing image data;
the positioning method of the whole chip mainly comprises the steps of highlighting the whole chip characteristics and removing unnecessary noise through the matched illumination imaging of a light source, extracting a bright area by adopting a binarization algorithm, further calculating the geometric information of the bright area, and performing affine transformation to correct the placement position of the chip, wherein the light source comprises a low-angle annular light source, a strip-shaped light source and a backlight light source;
the pin positioning method comprises the steps of scanning chip pins in a scanning line form, judging the specific placement position of the chip pins, providing a positioning method combining template matching and feature analysis, and further correcting the angle required to be converted to the correct position to realize the chip positioning and identifying function;
the image data processing flow method comprises the steps of carrying out color space transformation on an image to obtain average data of all images, and judging a preprocessed image effect graph to be selected by decision by setting a threshold range.
6. The IC lithographic character recognition and detection apparatus surface imaging method as claimed in claim 5, wherein: in the IC photoetching character recognition process, in order to accurately and quickly recognize characters, certain requirements are required on the recognized character postures, so that the character postures are corrected by further designing an algorithm, in a positive rectangular image, characters in some rectangular images are inverted postures, and a positioning algorithm is further designed to ensure that the characters are in a positive state;
the chip-based feature analysis and calculation method has two methods:
firstly, judging whether the current condition of the chip is an upright posture or not by screening and counting the number of chip pins;
secondly, forming a horizontal scanning line on the image of the rectangular area, scanning from top to bottom, and intersecting with the pin characteristics; judging whether the current condition of the chip is a positive attitude or not according to the intersection result;
the initial position of the scanning line and the working interval of the scanning line are set, which are the keys for ensuring accurate scanning and improving the running speed, and the specific threshold value needs to refer to the calculation result of the actual chip position.
7. The method of claim 6, wherein the imaging device comprises: in the method for positioning the pins by the horizontal scanning lines, the pins of the chip are scanned mainly in a form of forming the scanning lines, so that the specific placing positions of the pins of the chip are judged;
the initial position of the scanning line formation is to set an initial coordinate threshold value by calculating the central position of the characters on the surface of the chip, and further to set a spacing threshold value between the formed scanning lines, and the threshold value is set to have the main functions of:
firstly, the threshold setting can avoid the formation of unnecessary scanning lines, thereby saving the consumption of a computer memory and further saving the consumption of time;
secondly, due to the fact that the scanning lines are arranged at intervals, wrong judgment caused by different lengths of chip pins can be effectively avoided, meanwhile, time is saved for computer scanning, and the operation speed of an algorithm is further improved;
thirdly, in the scanning process, through the setting of threshold value, can judge the concrete position of chip pin, further can realize the function of pin measurement, and according to the scanning line and the pin and get the front and back contrast of the result of intersection, can judge whether the chip pin has the condition of lacking the pin, realize the defect detection function of chip pin.
8. The method of claim 7, wherein the imaging device comprises: the defect detection function of the chip pins is to mark corresponding results when corresponding image preprocessing is performed, so as to realize the effect of defect classification, such as:
firstly, when an optimal processing channel is selected, the distribution condition of the brightest pixels of the character is calculated, the calculated result can be adopted to judge whether the surface of the character is printed fully or lightly, if the character is printed fully, the value of the brightest pixels obtained by calculation is higher, otherwise, the obtained result value is lower;
secondly, when the connected domains of the number of the characters are counted, whether the characters printed by the current chip are lost or not can be judged according to the current counting result;
before obtaining the whole connected domain of the character, the whole character needs to be subjected to difference processing, the standard character and the character which needs to be identified currently are subjected to difference processing, and the difference result can be used for judging whether the current character is in the conditions of few strokes, multiple strokes or breakage and the like;
in the statistical classification of defect classification, all the defect classification is carried out by making corresponding algorithm marks in necessary image preprocessing for decision classification, and large-scale neural network training is not required, so that a better and faster classification and identification effect is achieved.
CN202210131721.0A 2022-02-14 2022-02-14 IC photoetching character recognition and detection device surface imaging system and method Pending CN114496860A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210131721.0A CN114496860A (en) 2022-02-14 2022-02-14 IC photoetching character recognition and detection device surface imaging system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210131721.0A CN114496860A (en) 2022-02-14 2022-02-14 IC photoetching character recognition and detection device surface imaging system and method

Publications (1)

Publication Number Publication Date
CN114496860A true CN114496860A (en) 2022-05-13

Family

ID=81479550

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210131721.0A Pending CN114496860A (en) 2022-02-14 2022-02-14 IC photoetching character recognition and detection device surface imaging system and method

Country Status (1)

Country Link
CN (1) CN114496860A (en)

Similar Documents

Publication Publication Date Title
CN109443480B (en) Water level scale positioning and water level measuring method based on image processing
US7505149B2 (en) Apparatus for surface inspection and method and apparatus for inspecting substrate
JP4946668B2 (en) Substrate position detection device and substrate position detection method
CN106960208A (en) A kind of instrument liquid crystal digital automatic segmentation and the method and system of identification
CN113109348B (en) Paddle image transfer printing defect identification method based on machine vision
CN115791822A (en) Visual detection algorithm and detection system for wafer surface defects
CN106501272B (en) Machine vision soldering tin positioning detection system
CN104331695B (en) A kind of circle marker symbol shape quality detection method of robust
CN107610085A (en) A kind of welding point defect detecting system based on computer vision
CN108709500B (en) Circuit board element positioning and matching method
JPH07260701A (en) Recognition method of area of inspection
CN114755243A (en) FPC connector surface crack detection method based on machine vision
CN112115948A (en) Chip surface character recognition method based on deep learning
CN112419260A (en) PCB character area defect detection method
CN107656387B (en) Liquid crystal display alignment offset defect detection method
CN114577805A (en) MiniLED backlight panel defect detection method and device
CN111929239A (en) AOI detection device and detection method for PCB part defects
CN113257182B (en) Lamp point position correction method and device in LED display screen correction process
CN111060518A (en) Stamping part defect identification method based on instance segmentation
CN112419225B (en) SOP type chip detection method and system based on pin segmentation
CN114496860A (en) IC photoetching character recognition and detection device surface imaging system and method
CN117198914A (en) BGA ball-mounting high-definition image acquisition device and image processing method
CN114092448B (en) Plug-in electrolytic capacitor mixed detection method based on deep learning
CN116433604A (en) Chip identification method and pickup system
CN114720473A (en) Defect detection method of vehicle-mounted screen backboard based on machine vision

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