CN116448780A

CN116448780A - Chip defect detection device, method and equipment

Info

Publication number: CN116448780A
Application number: CN202310232353.3A
Authority: CN
Inventors: 曹民; 林红; 胡秀文; 朱二路
Original assignee: Wuhan Optical Valley Excellence Technology Co ltd
Current assignee: Wuhan Optical Valley Excellence Technology Co ltd
Priority date: 2023-03-06
Filing date: 2023-03-06
Publication date: 2023-07-18

Abstract

The invention provides a chip defect detection device, method and equipment, and relates to the technical field of chip defect detection, wherein the device comprises: the device comprises a feeding mechanism, an image acquisition mechanism, a chip detection mechanism, a chip quality classification mechanism and a discharging mechanism; the feeding mechanism comprises a first feeding conveying assembly for transferring a tray to be tested, wherein the tray to be tested is used for storing chips to be tested; the image acquisition mechanism comprises an image acquisition assembly and an automatic adjustment assembly, and the image acquisition assembly is used for moving and acquiring images of the chip to be detected under the drive of the automatic adjustment assembly; the chip detection mechanism is used for carrying out defect detection and quality judgment on the chip to be detected based on the image of the chip to be detected; the chip quality classification mechanism is used for transferring chips with different quality categories to different blanking conveying components of the blanking mechanism according to the quality judgment result of the chip detection mechanism on the chip to be detected. The chip defect detection device, method and equipment provided by the invention can realize automatic detection of defects of various types of chips.

Description

Chip defect detection device, method and equipment

Technical Field

The present invention relates to the field of chip defect detection technologies, and in particular, to a device, a method, and an apparatus for detecting a chip defect.

Background

After the chip is processed, defect detection is needed, and defective products with defects such as breakage, foreign matters, welding and the like are screened out, so that the product quality is ensured.

Chip defects generally require manual sampling detection of batches of chip samples, and detection of samples requires manual visual inspection of products with a microscope to determine whether welding defects or damage exist. However, with the increase of the demand of chip products, the chip yield is larger and larger, the yield of the products is difficult to control by a manual sampling detection mode, and the manual detection efficiency is low, so that the detection rate and the delivery rate of the products are seriously affected.

Although related technologies related to automatic detection of chip defects exist at present, manual participation of operators is basically required, and automatic detection cannot be completely realized. Therefore, how to automatically detect defects of various types of chips is a problem that needs to be solved in the industry.

Disclosure of Invention

Aiming at the problems existing in the prior art, the invention provides a device, a method and equipment for detecting chip defects.

In a first aspect, the present invention provides a chip defect detection apparatus, including:

the device comprises a feeding mechanism, an image acquisition mechanism, a chip detection mechanism, a chip quality classification mechanism and a discharging mechanism;

The feeding mechanism comprises a first feeding conveying assembly and a second feeding conveying assembly, the first feeding conveying assembly is used for transporting a tray to be tested, the chip to be tested is stored in the tray to be tested, and the second feeding conveying assembly is used for transporting an empty tray;

the image acquisition mechanism comprises an image acquisition assembly and an automatic adjusting assembly, and the image acquisition assembly is used for moving under the drive of the automatic adjusting assembly and acquiring images of the chips to be detected stored on the tray to be detected;

the chip detection mechanism comprises a data processing unit, wherein the data processing unit comprises a data preprocessing subunit, a suspected target extraction subunit, a suspected target position fuzzy matching subunit, a target position information verification subunit, a chip defect judging subunit and a chip quality judging subunit, and the chip detection mechanism is used for carrying out defect detection and quality judgment on the chip to be detected based on the image of the chip to be detected, which is acquired by the image acquisition assembly, and based on the data processing unit;

the chip quality classification mechanism is used for transferring chips with different quality categories in the chip to be tested to the blanking mechanism according to the quality judgment result of the chip detection mechanism on the chip to be tested so as to finish quality classification of the chip to be tested;

The blanking mechanism comprises a first blanking conveying assembly, a second blanking conveying assembly and a third blanking conveying assembly, wherein the first blanking conveying assembly is used for conveying good product trays, the good product trays are used for storing chips with good quality judging results, the second blanking conveying assembly is used for conveying defective product trays, the defective product trays are used for storing chips with defective quality judging results, the third blanking conveying assembly is used for conveying first target trays, and the first target trays are empty trays which are produced after the chips to be detected stored on the trays to be detected are detected and sorted.

Optionally, according to the chip defect detection device provided by the invention, the data processing unit stores a standard template and a standard welding wiring diagram corresponding to various types of chips in advance;

the data preprocessing subunit is used for performing binarization processing on the image of the chip to be detected to obtain a binarized image, detecting characteristic points in the binarized image, matching the detected characteristic points with standard templates corresponding to the chips of multiple types one by one, determining a standard template with the maximum matching degree with the position information of each characteristic point as a target standard template based on a matching result, determining an average rotation angle of the characteristic points and the corresponding positions in the target standard template, and performing rotation operation on the binarized image according to the average rotation angle to obtain a rotation binary image;

The suspected target extraction subunit is configured to extract a suspected target in the rotation binary image according to the shape feature, the welding spot energy concentration feature and the linear feature of the welding wire of the chip to be detected, where the suspected target includes a suspected chip target, a suspected welding spot target, a suspected welding wire target, a suspected welding spot target and a suspected noise target;

the suspected target position fuzzy matching subunit is configured to perform fuzzy matching on position information of each suspected target and position information of an actual target in a target standard welding wiring diagram, so as to obtain a fuzzy matching result that the suspected target uniquely corresponds to the actual target, where the target standard welding wiring diagram is the standard welding wiring diagram corresponding to the target standard template, and the actual target is a target corresponding to the suspected target in the target standard welding wiring diagram;

the target position information verification subunit is configured to determine a first position threshold neighborhood with the suspected target as a center based on the fuzzy matching result, and determine actual position information of the suspected target based on position related information of other types of suspected targets in the first position threshold neighborhood;

The chip defect judging subunit is configured to perform position comparison on the actual position information of the suspected target and the position information of an actual target corresponding to the suspected target in the standard welding wiring diagram, obtain a position comparison result, and determine that the suspected target has a defect if the position comparison result does not meet a second position threshold;

the chip quality judging subunit is used for determining and marking the quality class of the chip to be detected as good products or defective products based on the judging result of the chip defect judging subunit.

Optionally, according to the chip defect detection device provided by the invention, the chip quality classification mechanism comprises a chip sorting manipulator and a tray monitoring assembly;

the chip sorting manipulator is used for acquiring the chip to be tested from the tray to be tested on the first feeding and conveying assembly and transferring the chip to be tested to the empty position of the tray to be good on the first discharging and conveying assembly according to the quality judging result of the chip to be tested by the chip detecting mechanism, acquiring the chip to be tested from the tray to be tested on the first feeding and conveying assembly and transferring the chip to be tested to the empty position of the tray to be bad on the second discharging and conveying assembly under the condition that the quality class of the chip to be tested is determined to be bad;

The material tray monitoring assembly comprises a good material tray monitoring assembly and a bad material tray monitoring assembly, wherein the good material tray monitoring assembly is located right above the first blanking conveying assembly and used for monitoring the number of chips and the empty positions in the good material tray on the first blanking conveying assembly, and the bad material tray monitoring assembly is located right above the second blanking conveying assembly and used for monitoring the number of chips and the empty positions in the bad material tray on the second blanking conveying assembly.

Optionally, according to the chip defect detection device provided by the invention, the chip quality classification mechanism further comprises a tray transferring manipulator;

the tray transfer manipulator is used for pushing a second target tray in the first feeding and conveying assembly to the first discharging and conveying assembly, pushing the first target tray in the first feeding and conveying assembly to the third discharging and conveying assembly, and the second target tray is the tray to be tested which is detected in the first feeding and conveying assembly and still stores good products.

Optionally, according to the chip defect detection device provided by the invention, the image acquisition assembly comprises a camera, a lens and a light source piece;

The lens is arranged on the camera, and the light source piece is positioned right below the lens and is spaced from the surface of the lens by a preset distance.

Optionally, according to the chip defect detection device provided by the invention, the automatic adjustment assembly comprises a first direction driving motor and a second direction driving motor;

the first direction driving motor is used for driving the image acquisition assembly to move along a first direction;

the second direction driving motor is used for driving the image acquisition assembly to move along a second direction.

Optionally, according to the chip defect detection device provided by the invention, the first feeding conveying assembly comprises a first conveying belt and a first feeding displacement encoder for driving the first conveying belt to move;

the second feeding conveying assembly comprises a second conveying belt and a second feeding displacement encoder for driving the second conveying belt to move.

Optionally, according to the chip defect detecting device provided by the invention, the first blanking conveying assembly comprises a third conveying belt and a first blanking displacement encoder for driving the third conveying belt to move;

the second blanking conveying assembly comprises a fourth conveying belt and a second blanking displacement encoder for driving the fourth conveying belt to move;

The third blanking conveying assembly comprises a fifth conveying belt and a third blanking displacement encoder used for driving the fifth conveying belt to move.

In a second aspect, the present invention provides a method for detecting a chip defect, including:

performing binarization processing on the image of the chip to be detected to obtain a binarized image;

detecting the characteristic points in the binarized image, matching the detected characteristic points with pre-stored standard templates corresponding to various types of chips one by one, and determining the standard template with the maximum matching degree with the position information of each characteristic point as a target standard template based on a matching result;

determining an average rotation angle of the feature points and corresponding positions in the target standard template, and performing rotation operation on the binarized image according to the average rotation angle to obtain a rotation binary image;

extracting suspected targets in the rotating binary image according to the shape characteristics, the welding spot energy concentration characteristics and the linear characteristics of the welding wires of the chip to be detected, wherein the suspected targets comprise a suspected chip target, a suspected welding spot target, a suspected welding wire target, a suspected welding pad target and a suspected noise target;

respectively carrying out fuzzy matching on the position information of each suspected target and the position information of an actual target in a target standard welding wiring diagram to obtain a fuzzy matching result of the suspected target and the actual target, wherein the target standard welding wiring diagram is the standard welding wiring diagram corresponding to the target standard template, and the actual target is a target corresponding to the suspected target in the target standard welding wiring diagram;

Determining a first position threshold neighborhood centering on the suspected target based on the fuzzy matching result, and determining actual position information of the suspected target based on position association information of other types of suspected targets in the first position threshold neighborhood;

and comparing the actual position information of the suspected target with the position information of the actual target corresponding to the suspected target in the standard welding wiring diagram to obtain a position comparison result, and determining that the suspected target has a defect under the condition that the position comparison result does not meet a second position threshold value.

In a third aspect, the present invention also provides a chip defect detection apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the chip defect detection method according to the second aspect when executing the program.

According to the chip defect detection device, the chip defect detection method and the chip defect detection equipment, the first feeding conveying component in the feeding mechanism is used for conveying the to-be-detected tray stored with the to-be-detected chips, the image acquisition component in the image acquisition mechanism is driven by the automatic adjustment component to move and acquire images of the to-be-detected chips stored on the to-be-detected tray, then the chip detection mechanism is used for carrying out defect detection and quality judgment on the to-be-detected chips based on the images of the to-be-detected chips acquired by the image acquisition component, and the chip quality classification mechanism is used for conveying the chips with different quality categories in the to-be-detected chips to the first blanking conveying component and the second blanking conveying component in the blanking mechanism according to the quality judgment result of the chip detection mechanism, so that the quality classification of the to-be-detected chips is finished, the integrated arrangement of loading, detection, sorting and blanking is realized, and automatic detection of defects of various types of chips can be effectively realized.

Drawings

In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic diagram of a chip defect detecting device according to the present invention;

fig. 2 is a schematic flow chart of a chip defect detection method provided by the invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

It should be noted that, in the description of the present invention, the terms "first," "second," and the like are used for distinguishing between similar objects and not for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged, as appropriate, such that embodiments of the present invention may be implemented in sequences other than those illustrated or described herein, and that the objects identified by "first," "second," etc. are generally of a type, and are not limited to the number of objects, such as the first object may be one or more.

The chip defect detection device, method and equipment provided by the invention are exemplarily described below with reference to the accompanying drawings.

Fig. 1 is a schematic structural diagram of a chip defect detecting device according to the present invention, as shown in fig. 1, the device includes:

the chip quality sorting device comprises a feeding mechanism 100, an image acquisition mechanism 110, a chip detection mechanism 120, a chip quality sorting mechanism 130 and a discharging mechanism 140;

the feeding mechanism 100 comprises a first feeding and conveying assembly and a second feeding and conveying assembly, wherein the first feeding and conveying assembly is used for transferring a tray to be tested, in which chips to be tested are stored, and the second feeding and conveying assembly is used for transferring an empty tray;

the image acquisition mechanism 110 comprises an image acquisition component and an automatic adjustment component, wherein the image acquisition component is used for moving under the drive of the automatic adjustment component and acquiring images of the chips to be tested stored on the tray to be tested;

the chip detection mechanism 120 includes a data processing unit, where the data processing unit includes a data preprocessing subunit, a suspected target extraction subunit, a suspected target position fuzzy matching subunit, a target position information verification subunit, a chip defect determination subunit, and a chip quality determination subunit, and the chip detection mechanism is configured to perform defect detection and quality determination on the chip to be detected based on the image of the chip to be detected acquired by the image acquisition component and based on the data processing unit;

The chip quality classification mechanism 130 is configured to transfer chips with different quality categories in the chip to be tested to the blanking mechanism according to the quality determination result of the chip detection mechanism 120 on the chip to be tested, so as to complete quality classification of the chip to be tested;

the blanking mechanism 140 comprises a first blanking conveying assembly, a second blanking conveying assembly and a third blanking conveying assembly, the first blanking conveying assembly is used for conveying good product trays, the good product trays are used for storing chips with good quality judging results, the second blanking conveying assembly is used for conveying defective product trays, the defective product trays are used for storing chips with defective quality judging results, the third blanking conveying assembly is used for conveying first target trays, and the first target trays are empty trays which are produced after the chips to be detected stored on the trays to be detected are detected and sorted.

Specifically, in order to overcome the defect that the existing chip defect detection method basically needs manual participation of operators and fails to completely realize automatic detection, the invention transfers the to-be-detected tray stored with the to-be-detected chips through the first feeding and conveying assembly in the feeding mechanism 100, so that the image acquisition assembly in the image acquisition mechanism 110 moves under the drive of the automatic adjusting assembly and acquires the images of the to-be-detected chips stored on the to-be-detected tray, then the chip detection mechanism 120 performs defect detection and quality judgment on the to-be-detected chips based on the images of the to-be-detected chips acquired by the image acquisition assembly, and the chip quality classification mechanism 130 transfers the chips with different quality categories in the to-be-detected chips to the first discharging and conveying assembly and the second discharging and conveying assembly in the discharging mechanism 140 according to the quality judgment result of the chip detection mechanism, thereby completing quality classification of the to-be-detected chips, realizing integrated setting of feeding, detection, sorting and discharging, and further effectively realizing automatic detection on defects of various types of chips.

Optionally, in an embodiment of the present invention, the feeding mechanism 100 may include a first feeding conveying assembly and a second feeding conveying assembly, where the first feeding conveying assembly is used for transporting a tray to be tested storing chips to be tested, and the second feeding conveying assembly is used for transporting an empty tray, where the empty tray may be used for storing detected defective chips.

Alternatively, in an embodiment of the present invention, the image capturing mechanism 110 may include an image capturing component and an automatic adjustment component, where the automatic adjustment component may drive the image capturing component to move in space, and the image capturing component is used to move above each chip to be tested under the drive of the automatic adjustment component, and simultaneously automatically focus and capture the image of the chip to be tested stored on the tray to be tested.

Optionally, in an embodiment of the present invention, the chip detection mechanism 120 may be configured to perform defect detection and quality determination on the chip to be detected based on the image of the chip to be detected acquired by the image acquisition component, so as to obtain a quality determination result of the chip to be detected.

Optionally, in the embodiment of the present invention, the chip quality classification mechanism 130 may be configured to transfer chips with different quality categories in the chip to be tested to different blanking conveying components of the blanking mechanism according to the quality determination result of the chip to be tested by the chip detection mechanism 120, so as to complete quality classification of the chip to be tested.

Optionally, in an embodiment of the present invention, the blanking mechanism 140 may include a first blanking conveying assembly, a second blanking conveying assembly and a third blanking conveying assembly, where the first blanking conveying assembly is used for conveying good product trays, the good product trays are used for storing chips with good quality determination results conveyed by the chip quality classification mechanism 130, the second blanking conveying assembly is used for conveying defective product trays, the defective product trays are used for storing chips with defective quality determination results conveyed by the chip quality classification mechanism 130, the third blanking conveying assembly is used for conveying first target trays, and the first target trays are empty trays generated after the chips to be tested stored on the trays to be tested are detected and sorted.

Optionally, in an embodiment of the present invention, the first blanking conveying assembly and the second blanking conveying assembly are further used for conveying chips with different quality classes to a next station.

According to the chip defect detection device provided by the invention, the first feeding conveying component in the feeding mechanism is used for conveying the to-be-detected tray storing the to-be-detected chips, so that the image acquisition component in the image acquisition mechanism moves under the drive of the automatic adjustment component and acquires the images of the to-be-detected chips stored on the to-be-detected tray, then the chip detection mechanism is used for carrying out defect detection and quality judgment on the to-be-detected chips based on the images of the to-be-detected chips acquired by the image acquisition component, and the chip quality classification mechanism is used for conveying the chips with different quality categories in the to-be-detected chips to the first blanking conveying component and the second blanking conveying component in the blanking mechanism according to the quality judgment result of the chip detection mechanism, thereby finishing quality classification of the to-be-detected chips, realizing integrated setting of feeding, detection, sorting and blanking, and automatic detection of defects of various types of chips can be effectively realized.

Optionally, the chip detection mechanism 120 includes a data processing unit, where the data processing unit stores in advance a standard template and a standard welding wiring diagram corresponding to the chips of multiple types respectively;

the data processing unit comprises a data preprocessing subunit, a suspected target extraction subunit, a suspected target position fuzzy matching subunit, a target position information verification subunit, a chip defect judging subunit and a chip quality judging subunit;

Specifically, in the embodiment of the present invention, the chip detection mechanism 120 includes a data processing unit, in which a standard template and a standard solder pattern corresponding to a plurality of types of chips are stored in advance, and the data processing unit includes a data preprocessing subunit, a suspected target extraction subunit, a suspected target location fuzzy matching subunit, a target location information verification subunit, a chip defect determination subunit, and a chip quality determination subunit, where: (1) The data preprocessing subunit can be used for carrying out binarization processing on an image of a chip to be detected to obtain a binarized image, detecting characteristic points in the binarized image, matching the detected characteristic points with standard templates corresponding to various types of chips one by one, determining the standard template with the maximum matching degree with the position information of each characteristic point as a target standard template based on a matching result, determining an average rotation angle of the characteristic point and the corresponding position in the target standard template, and carrying out rotation operation on the binarized image according to the average rotation angle to obtain a rotation binarized image; (2) The suspected target extraction subunit may be configured to extract a suspected target in the rotation binary image according to the shape feature, the welding spot energy concentration feature, and the linear feature of the welding wire of the chip to be detected, where the suspected target may include a suspected chip target, a suspected welding spot target, a suspected welding wire target, a suspected welding pad target, a suspected noise target, and the like; (3) The suspected target position fuzzy matching subunit can be used for respectively carrying out fuzzy matching on the position information of each suspected target and the position information of an actual target in a target standard welding wiring diagram to obtain a fuzzy matching result uniquely corresponding to the suspected target and the actual target, wherein the target standard welding wiring diagram is a standard welding wiring diagram corresponding to a target standard template, and the actual target is a target corresponding to the suspected target in the target standard welding wiring diagram; (4) The target position information verification subunit may be configured to determine a first position threshold neighborhood with the suspected target as a center based on the fuzzy matching result, and determine actual position information of the suspected target based on position related information of other types of suspected targets in the first position threshold neighborhood; (5) The chip defect judging subunit can be used for comparing the actual position information of the suspected target with the position information of the actual target corresponding to the suspected target in the standard welding wiring diagram to obtain a position comparison result, and determining that the suspected target has a defect under the condition that the position comparison result does not meet a second position threshold value; (6) The chip quality judging subunit may be configured to determine and mark the quality class of the chip to be tested as good or bad based on the judging result of the chip defect judging subunit.

It should be noted that, in the embodiment of the present invention, the functions of the data preprocessing subunit include binarization processing of the chip image, feature point detection in the binarized image, and template matching. The feature points may be edge corner points, welding points, and pad center points of the chip in the binarized image, which is not particularly limited in the embodiment of the present invention. The template matching refers to that the detected characteristic points can be matched with standard templates corresponding to various types of chips stored in a template library one by one, the standard template with the largest matching degree is determined to be a target standard template, then the average rotation angle of the corresponding position in the target standard template is calculated according to the matched characteristic points, and further the binarization image is rotated according to the average rotation angle, so that a rotation binary image is obtained.

It should be noted that, considering that the chips to be tested are randomly placed in the tray to be tested, in the embodiment of the invention, the binary image needs to be rotated, and then the binary image of the chips to be tested is adjusted to be approximately the same angle as the standard template after the binary image is rotated, and if the chips or welding spots are detected later, the angle does not need to be introduced into calculation.

It can be understood that in the embodiment of the present invention, the suspected target position fuzzy matching subunit performs fuzzy position matching on the extracted suspected targets, by using the position information of each extracted single suspected target and combining the position information of the actual target in the target standard welding wiring diagram, on the single suspected target one by one, so as to obtain a corresponding fuzzy matching result.

It may be appreciated that the target location information verification subunit may determine, according to a fuzzy matching result uniquely corresponding to the suspected target and the actual target, a first location threshold neighborhood centered on each suspected target, and further determine the actual location information of each suspected target based on location association information of other types of suspected targets in the first location threshold neighborhood.

The first position threshold neighborhood may be a fixed value or an adaptive value generated by a minimum bounding rectangle section of the suspected target, the minimum bounding rectangle is taken as a central section, each boundary of the minimum bounding rectangle is taken as a starting point, the threshold T is taken as a first position threshold extension bounding rectangle boundary, the extended section is taken as a first position threshold neighborhood, and further, the threshold T may be a fixed value or an adaptive value generated by the minimum bounding rectangle section of the suspected target.

It may be appreciated that in the embodiment of the present invention, the chip defect determining subunit may determine that the suspected target has a defect when determining that the second position threshold is not satisfied by performing the position threshold determination on the actual position in the verification of the position information of the suspected target and the standard position in the standard welding wiring diagram.

It should be noted that the second position threshold may be adaptively set based on practical applications, which is not specifically limited in the embodiment of the present invention.

It can be understood that, in the embodiment of the present invention, the chip quality determining subunit may perform quality classification on the currently detected chip according to the determination result of the chip defect determining subunit, and the defective chip determines that the quality classification result is defective, or determines that the quality classification result is good.

It should be noted that, because the existing automatic detection method for chip defects is difficult to realize comprehensive detection for defects of finished products of chips of various types, the embodiment of the invention can realize detection for the chip defects by a fuzzy matching method, and can adaptively detect defects such as welding spots, welding lines, misplacement welding, missing welding and the like in chips of various types, thereby realizing comprehensive automatic detection for chips of various types.

Optionally, the chip quality sorting mechanism 130 includes a chip sorting robot and a tray monitoring assembly;

the chip sorting manipulator is configured to obtain, according to a quality determination result of the chip detection mechanism 120 on the chip to be detected, the chip to be detected from the tray to be detected on the first feeding and conveying assembly and transfer the chip to be detected to an empty position of the tray to be detected on the first discharging and conveying assembly when the quality category of the chip to be detected is determined to be a defective product, obtain the chip to be detected from the tray to be detected on the first feeding and conveying assembly and transfer the chip to be detected to an empty position of the defective product tray on the second discharging and conveying assembly;

Specifically, in the embodiment of the present invention, the chip quality classification mechanism 130 may include a chip sorting mechanical arm and a tray monitoring component, where the chip sorting mechanical arm may be configured to obtain, according to a quality determination result of a chip to be tested by the chip detection mechanism 120, the chip to be tested from a tray to be tested on the first feeding and conveying component and transfer the obtained chip to an empty position of a good tray on the first blanking and conveying component, and obtain, under a condition that the quality category of the chip to be tested is determined to be a defective product, the chip to be tested from the tray to be tested on the first feeding and conveying component and transfer the obtained chip to an empty position of a defective tray on the second blanking and conveying component; the charging tray monitoring assembly comprises a good product charging tray monitoring assembly and a defective product charging tray monitoring assembly, wherein the good product charging tray monitoring assembly is located right above the first blanking conveying assembly and can be used for monitoring the number of chips and the vacant positions in the good product charging tray on the first blanking conveying assembly, and the defective product charging tray monitoring assembly is located right above the second blanking conveying assembly and can be used for monitoring the number of chips and the vacant positions in the defective product charging tray on the second blanking conveying assembly.

It may be appreciated that, in the embodiment of the present invention, the chip sorting manipulator may sort the current detected chips according to the quality determination result of the chip detection mechanism 120, if the detected chips are good, the chip sorting manipulator grabs chips with the current quality determination result being good from the trays to be detected on the first feeding conveyor, and sequentially stores the chips to the empty positions in the trays to be detected on the first discharging conveyor, if the chips are bad, the chip sorting manipulator grabs chips with the current quality determination result being bad from the trays to be detected on the first feeding conveyor, and sequentially stores the chips to the empty positions in the trays to be detected on the second discharging conveyor, wherein the empty positions in the trays to be detected on the first discharging conveyor and the empty positions in the trays to be detected on the second discharging conveyor are obtained by the tray monitoring component.

Optionally, in the embodiment of the present invention, the tray monitoring assembly includes a good tray monitoring assembly and a bad tray monitoring assembly, where the good tray monitoring assembly may be located directly above the first blanking conveying assembly and is composed of a camera, a lens and a monitoring device, and is used for monitoring the number of chips and the empty positions in the good tray in real time; the inferior quality material tray monitoring component can be positioned right above the second blanking conveying device and consists of a camera, a lens and monitoring equipment and is used for monitoring the number of chips in the inferior quality material tray and the vacant position in real time.

According to the chip sorting machine, the chip sorting mechanical arm and the tray monitoring assembly in the chip quality sorting mechanism are used for automatically sorting good chips and defective chips.

Optionally, the chip quality classification mechanism 130 further includes a tray transfer robot;

Specifically, in the embodiment of the present invention, the chip quality classification mechanism 130 includes a tray transfer manipulator in addition to the chip sorting manipulator and the tray monitoring component, where the tray transfer manipulator is configured to push the second target tray in the first feeding and conveying component to the first discharging and conveying component, and push the first target tray in the first feeding and conveying component to the third discharging and conveying component, where the second target tray is a tray to be tested that has been detected in the first feeding and conveying component and still has good products stored therein.

It can be appreciated that in the embodiment of the invention, the tray transferring manipulator can push the trays in the feeding mechanism to the discharging mechanism in parallel so as to realize recycling of the trays.

Optionally, in an embodiment of the present invention, the chip quality sorting mechanism 130 may further include a controller, where the controller may be configured to control the co-operation of the chip sorting manipulator, the tray transferring manipulator, and the tray monitoring assembly, receive the quality determination result of the chip detecting mechanism 120, control the chip sorting manipulator to grasp the chip through an instruction, and control a series of operations such as the tray transferring manipulator to transfer the tray through an instruction, so as to implement automatic sorting of the chip.

Optionally, the image acquisition assembly comprises a camera, a lens and a light source component;

Specifically, in an embodiment of the present invention, the image capturing assembly in the image capturing mechanism 110 may include a camera, a lens, and a light source member, wherein the lens is mounted on the camera, and the light source member is located directly below the lens and spaced apart from the surface of the lens by a preset distance.

Alternatively, the preset distance between the light source element and the surface of the lens may be adaptively set based on practical applications, which is not particularly limited in the embodiment of the present invention.

Optionally, the automatic adjustment assembly includes a first direction drive motor and a second direction drive motor;

Specifically, in an embodiment of the present invention, the automatic adjustment component in the image capturing mechanism 110 may include a first direction driving motor for driving the image capturing component to move in the first direction and a second direction driving motor for driving the image capturing component to move in the second direction.

It can be appreciated that in the embodiment of the present invention, the image acquisition assembly can be moved to a suitable position by driving the first direction driving motor and the second direction driving motor, so as to perform image acquisition on the chip to be measured in the tray to be measured on the first feeding and conveying assembly.

Specifically, in the embodiment of the present invention, the image capturing mechanism 110 may be located above the first feeding and conveying assembly, and includes an image capturing assembly composed of a camera, a lens and a light source component, and an automatic adjusting assembly for controlling the image capturing assembly to move in space, where the lens is fixedly assembled on the camera, the light source component is located under the lens and is fixedly installed on the surface of the lens at a preset distance, and the automatic adjusting assembly includes a first direction driving motor and a second direction driving motor, where the first direction driving motor drives the image capturing assembly to move in a first direction with high precision, and the second direction driving motor drives the image capturing assembly to move in a second direction with high precision.

Optionally, the first feeding conveying assembly comprises a first conveying belt and a first feeding displacement encoder for driving the first conveying belt to move;

Specifically, in the embodiment of the invention, the first feeding conveying assembly can comprise a first conveying belt and a first feeding displacement encoder for driving the first conveying belt to move with high precision, and the second feeding conveying assembly comprises a second conveying belt and a second feeding displacement encoder for driving the second conveying belt to move with high precision.

Optionally, the first blanking conveying assembly includes a third conveyor belt and a first blanking displacement encoder for driving the third conveyor belt to move;

Specifically, in an embodiment of the present invention, the first discharging conveyor assembly may include a third conveyor belt and a first discharging displacement encoder for driving the third conveyor belt to move with high precision, the second discharging conveyor assembly may include a fourth conveyor belt and a second discharging displacement encoder for driving the fourth conveyor belt to move with high precision, and the third discharging conveyor assembly may include a fifth conveyor belt and a third discharging displacement encoder for driving the fifth conveyor belt to move with high precision.

The chip defect detection device provided by the invention is described below through a specific embodiment.

It should be noted that, in the embodiment of the present invention, the feeding mechanism includes two sets of conveying components (a first feeding conveying component and a second feeding conveying component), and each set of conveying components is composed of a conveying belt and a high-precision displacement encoder. The tray to be tested is transported to the first feeding conveying assembly by the production line of chip products, and under the control of the high-precision first feeding conveying encoder, the first feeding conveying assembly transports the tray to be tested to the position right below the image acquisition mechanism.

The automatic adjusting component in the image acquisition mechanism utilizes a motor in a first direction (a first direction driving motor) and a motor in a second direction (a second direction driving motor) to place the image acquisition component in a zero position, namely the left upper corner position of the charging tray. The camera in the image acquisition component performs automatic zooming to find the clear point position of the chip to be detected in the visual field, acquires a frame of image of the current chip to be detected and transmits the frame of image to the chip detection mechanism. After the first time threshold interval, the automatic adjusting component controls the image acquisition mechanism to move above the next chip to be tested in two directions, and the next image acquisition operation is repeated. The first time threshold may be adaptively set based on practical applications, which is not specifically limited in the embodiment of the present invention.

The chip detection mechanism detects chip defects according to the chip images acquired by the image acquisition mechanism. Firstly, preprocessing a chip image, and selecting a self-adaptive first threshold value to binarize the chip image, wherein the first threshold value can be the mean value of gray values of a full graph multiplied by a first coefficient K to obtain a binarized image; then, detecting characteristic points of the binarized image, wherein the characteristic points can be edge corner points, welding points and welding disc center points of the chip, matching the characteristic points with standard templates in a chip template library one by one, selecting a chip template with the largest matching degree of the position information of the characteristic points as a template of a current chip to be detected, acquiring and recording a standard welding wiring diagram corresponding to the current template, calculating an average rotation angle theta of the characteristic points and corresponding positions in the template, and rotating the binarized image by the theta to obtain a rotating binary image; further performing shape detection in the rotated binary image, such as square and round, marking the target meeting the first area threshold as a suspected chip target, marking the target for detecting the linear feature in the rotated binary image as a suspected bonding wire target, detecting the point with concentrated energy as a suspected bonding spot target, and detecting other square and round targets meeting the second area threshold in a first neighborhood near the suspected chip target as bonding pad targets.

Further, the suspected targets detected in the rotation binary images are compared with the neighborhood meeting the first position threshold in the current standard template one by one, and fuzzy matching based on the position information is carried out. And comprehensively judging whether the current suspected target is a real chip, a welding point, a welding line or a welding pad according to the position information of each pair of actual targets and the suspected targets in fuzzy matching and the position associated information of other types of suspected targets in the current area, reserving the real suspected targets, and deleting the non-real suspected targets. Meanwhile, the finally reserved suspected targets are compared with the same target at the current position in the standard template, the targets meeting the threshold value of the second position are marked as normal, and the targets not meeting the threshold value are marked as defects.

Finally, after all suspected targets are subjected to target position information verification, marking the positions of other targets which are not matched in the template, such as welding lines, welding spots and welding pads, in a rotation binary image as defects, and possibly, missing welding; judging whether the target positions in the current standard template are correctly marked or not, judging the current chip according to the marking result, marking the chip with the defect mark as a defective product, and marking the chip without the defect mark as a good product; and then the quality judgment result mark of the current chip to be tested is sent to the chip sorting manipulator.

The charging tray monitoring assembly comprises a good product charging tray monitoring assembly and a defective product charging tray monitoring assembly. The camera of the good product tray monitoring assembly collects images of the good product chip trays at a certain sampling interval, processes the collected images, determines information of the good product trays and position information of the next good product chip to be placed on the good product trays, and sends marks of whether to transfer the trays to the first blanking conveying assembly; meanwhile, the defective material tray monitoring component also determines the information of the defective material tray and the position information of the next defective material chip to be placed on the defective material tray in the same mode, and sends the mark of whether to transfer the material tray to the second blanking conveying component.

The chip sorting manipulator is used for grabbing a currently detected chip and moving the currently detected chip to the upper part of a good product tray on the first blanking platform according to the received sorting marks, and placing the good product chip according to the empty position information of the good product provided by the good product tray monitoring assembly if the chip sorting manipulator is a defective product, and placing the defective product chip according to the empty position information of the defective product provided by the defective product tray monitoring assembly if the chip sorting manipulator is a defective product.

If the good product tray monitoring assembly does not monitor the good product tray information, the chip sorting manipulator does not grasp the good product chips, the chips detected as defective products in the current to-be-detected tray are grasped to the defective product tray, the good product chips are temporarily stored to the to-be-detected tray until all the current to-be-detected chips are detected, and the tray transferring manipulator pushes the detected current tray on the first feeding and conveying assembly to be right below the good product tray monitoring assembly on the first discharging and conveying assembly in parallel.

If the good product charging tray monitoring component monitors the information of the good product charging tray, and the empty position is not available in the good product charging tray, the good product charging tray monitoring component sends a signal to the first blanking conveying component, and the charging tray with the good product is transported to the next station. Similarly, if no empty position exists in the defective material tray, the second blanking conveying assembly conveys the material tray filled with the defective material to the next station. If all the current chips to be detected are detected, and the empty position information is still monitored by the good product tray monitoring assembly, the current trays to be detected are characterized by being emptied, the trays to be detected on the first feeding conveying assembly are pushed to the third discharging conveying assembly in parallel by the tray transferring manipulator, and are transferred to the next station, so that the chip defect detection is completed.

The chip defect detection method provided by the invention is described below, and the chip defect detection method described below can be applied to the chip defect detection device described above.

Fig. 2 is a schematic flow chart of a method for detecting a chip defect according to the present invention, as shown in fig. 2, the method includes:

step 200, performing binarization processing on an image of a chip to be tested to obtain a binarized image;

step 210, detecting feature points in the binarized image, matching the detected feature points with pre-stored standard templates corresponding to various types of chips one by one, and determining the standard template with the maximum matching degree with the position information of each feature point as a target standard template based on a matching result;

step 220, determining an average rotation angle of the feature points and corresponding positions in the target standard template, and performing rotation operation on the binarized image according to the average rotation angle to obtain a rotation binary image;

step 230, extracting suspected targets in the rotation binary image according to the shape features, the welding spot energy concentration features and the linear features of the welding wires of the chip to be detected, wherein the suspected targets comprise a suspected chip target, a suspected welding spot target, a suspected welding wire target, a suspected welding pad target and a suspected noise target;

Step 240, performing fuzzy matching on the position information of each suspected target and the position information of an actual target in a target standard welding wiring diagram, so as to obtain a fuzzy matching result of the suspected target and the actual target, wherein the target standard welding wiring diagram is the standard welding wiring diagram corresponding to the target standard template, and the actual target is a target corresponding to the suspected target in the target standard welding wiring diagram;

step 250, determining a first position threshold neighborhood centering on the suspected target based on the fuzzy matching result, and determining actual position information of the suspected target based on position related information of other types of suspected targets in the first position threshold neighborhood;

and 260, comparing the actual position information of the suspected target with the position information of the actual target corresponding to the suspected target in the standard welding wiring diagram to obtain a position comparison result, and determining that the suspected target has a defect under the condition that the position comparison result does not meet a second position threshold value.

Specifically, in the embodiment of the invention, an image of a chip to be tested can be obtained first, and binarization processing is performed on the image of the chip to be tested to obtain a binarized image; then, detecting characteristic points in the obtained binarized image, matching the detected characteristic points with the standard templates corresponding to the pre-stored chips of various types one by one, and determining the standard template with the maximum matching degree with the position information of each characteristic point as a target standard template based on a matching result; further determining an average rotation angle of the feature points and corresponding positions in the target standard template, and performing rotation operation on the obtained binary image according to the average rotation angle to obtain a rotation binary image; further, according to the shape characteristics of the chip to be detected, the welding spot energy concentration characteristics and the linear characteristics of the welding wire, extracting suspected targets in the rotating binary image, wherein the suspected targets comprise a suspected chip target, a suspected welding spot target, a suspected welding wire target, a suspected welding pad target, a suspected noise target and the like; then, fuzzy matching is carried out on the position information of each suspected target and the position information of an actual target in a target standard welding wiring diagram respectively, so as to obtain a fuzzy matching result which is unique between the suspected target and the actual target, wherein the target standard welding wiring diagram is a standard welding wiring diagram corresponding to a target standard template, and the actual target is a target corresponding to the suspected target in the target standard welding wiring diagram; further, based on the obtained fuzzy matching result, determining a first position threshold neighborhood centering on each suspected target, and based on position association information of other types of suspected targets in the first position threshold neighborhood, determining actual position information of each suspected target; and finally, carrying out position comparison on the actual position information of the suspected target and the position information of the actual target corresponding to the suspected target in the standard welding wiring diagram to obtain a position comparison result, and determining that the suspected target has defects under the condition that the position comparison result does not meet a second position threshold value.

According to the chip defect detection method provided by the invention, the detection of the chip defects is realized by utilizing the fuzzy matching method based on the binary image of the chip to be detected and the standard templates corresponding to the chips of the plurality of types stored in the template library, so that the defects of welding spots, welding wires, misplacement welding, missed welding and the like in the chips of the plurality of types can be detected in a self-adaptive manner, and the comprehensive automatic detection of the chips of the plurality of types is realized.

The present invention also provides a chip defect detecting apparatus, which may include: a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor implementing the chip defect detection method provided by the above method when executing the program, the method comprising:

In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method of detecting a chip defect provided by the above methods, the method comprising:

In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the above-provided chip defect detection methods, the method comprising:

The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A chip defect detection apparatus, comprising:

2. The chip defect detecting apparatus according to claim 1, wherein the data processing unit stores in advance a standard template and a standard solder pattern respectively corresponding to the plurality of types of chips;

3. The chip defect detection apparatus of claim 1, wherein the chip quality sorting mechanism comprises a chip sorting robot and a tray monitoring assembly;

4. The chip defect detection apparatus according to claim 3, wherein the chip quality classification mechanism further comprises a tray transfer robot;

5. The chip defect detection apparatus of claim 1, wherein the image acquisition assembly comprises a camera, a lens, and a light source;

6. The chip defect detection apparatus of claim 1, wherein the automatic adjustment assembly comprises a first direction drive motor and a second direction drive motor;

7. The chip defect detection apparatus of claim 1, wherein the first feed conveyor assembly comprises a first conveyor belt and a first feed displacement encoder for driving the first conveyor belt to move;

8. The chip defect inspection apparatus of claim 1, wherein the first blanking transfer assembly includes a third transfer belt and a first blanking displacement encoder for driving the third transfer belt to move;

9. A method for detecting a chip defect, comprising:

10. A chip defect detection apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the chip defect detection method of claim 9 when executing the program.