CN116609359A - Chip bonding wire detection method and device, electronic equipment and storage medium - Google Patents

Abstract

The present application relates to a chip bonding wire detection method, device, electronic equipment and storage medium. The method includes: receiving a chip image uploaded by an image acquisition device; determining whether the chip corresponding to the chip image is a target chip according to the chip image; if the chip image The corresponding chip is the target chip, then the target chip is controlled to enter the wire bonding machine for wire bonding to generate the target chip after the wire bonding; the machine with the amplification effect is started to collect the image of the target chip after the wire bonding; the receiver has the zoom effect According to the image collected by the machine, determine whether the target chip after wire bonding has abnormal wire bonding; if the target chip after wire bonding does not have abnormal wire bonding, start the push-pull machine for pulling force detection; receive the pull force of the push-pull machine As for the detection result, according to the tension detection result, it is determined whether the wire bonding of the target chip after the wire bonding is firm; if the wire bonding of the target chip after the wire bonding is firm, then it is determined that the target chip after the wire bonding is qualified.

Description

A chip bonding wire detection method, device, electronic equipment and storage medium

technical field

The present application relates to the field of chip technology, in particular to a chip bonding wire detection method, device, electronic equipment and storage medium.

Background technique

With the advancement of science and technology, the degree of automation in the modern production process has been continuously improved, the production capacity of packaged ICs has been greatly improved, and the requirements for its testing have also been continuously improved.

At present, the detection of IC is more manual detection, but the traditional manual detection requires a lot of manpower and material resources to train professional detection personnel, and the controllability of manual detection is poor. Visual fatigue, the accuracy of IC detection can not be controlled.

Contents of the invention

The present application provides a chip bonding wire detection method, device, electronic equipment and storage medium to solve the above technical problems.

In a first aspect, the present application provides a chip bonding wire detection method, including:

receiving the chip image uploaded by the image acquisition device;

According to the chip image, determine whether the chip corresponding to the chip image is a target chip;

If the chip corresponding to the chip image is a target chip, then controlling the target chip to enter the wire bonding machine for wire bonding to generate a target chip after wire bonding;

Start the machine with the amplification effect to collect the image of the target chip after the wire bonding on the target chip after the wire bonding;

Receiving the image collected by the machine with the magnification effect, and determining whether the target chip after the wire bonding has abnormal wire bonding according to the image;

If there is no abnormal wire bonding in the target chip after the wire bonding, then start the push-pull machine to perform tension detection;

receiving the pulling force detection result of the push-pull machine, and determining whether the wire bonding of the target chip after the wire bonding is firm according to the pulling force detection result;

If the wire bonding of the target chip after wire bonding is firm, it is determined that the target chip after wire bonding is qualified.

Through the solution provided by this application, before the target chip is bonded, the chip is detected by the image acquisition device to determine the target chip, and the defective chips are minimized to be bonded, resulting in waste of resources. In addition, after welding, use a machine with amplifying effect to determine whether there is a problem with the welding wire. If there is no problem, use a push-pull machine to detect the pulling force of the welding wire. On the basis that there is no problem with the bonding wire, ensure that there is no problem with the tension of the bonding wire, thereby ensuring that the chip bonding wire is qualified and reducing the occurrence of defective products.

Optionally, the determining whether the chip corresponding to the chip image is a target chip according to the chip image includes:

Determine the model of the chip corresponding to the chip image through the chip image;

Determine a preset standard image according to the model; the preset standard image is an image corresponding to the model without any problem;

Matching the chip image with the preset standard image to determine the matching degree;

The matching degree is compared with a preset matching threshold, and if the matching degree is higher than the preset matching threshold, the chip corresponding to the chip image is used as a target chip.

Through the method provided in this embodiment, when determining the target chip, the model of the chip can be determined first through image recognition, thereby calling the corresponding preset standard image, and then comparing to determine the matching degree. And set a preset matching threshold corresponding to each model, and when the matching degree is higher than the preset matching threshold, this chip can be used as the target chip. Determining the target chip through image recognition can reduce false detection problems that may be caused by human identification, reduce cost losses, and improve chip utilization.

Optionally, the matching degree includes a color matching degree, and matching the chip image with the preset standard image to determine the matching degree includes:

performing grid division on the chip image and the preset standard image according to a preset method;

Comparing the screen color of each grid of the chip image with the screen color of each grid of the preset standard image to determine whether there are grids with inconsistent colors;

If there are grids with inconsistent colors, the color matching degree is determined according to the number of grids with inconsistent colors.

Through the method provided by this embodiment, the chip image can be identified and distinguished by the color of the image, so as to determine the difference between the chip and the standard chip, and then determine whether this chip can be used as a target chip. Through this image recognition method, the matching degree of the chip can be determined as comprehensively as possible, and the error caused by human detection can be reduced.

Optionally, the determining whether the target chip after the wire bonding has abnormal wire bonding according to the image includes:

Identify the leads of the image, and determine whether the number of bonding wires of the target chip after the bonding wires is correct;

If the number of bonding wires of the target chip after the wire bonding is correct, extract the bonding wire path characteristics of the leads, and determine whether there is an abnormality in the bonding wire path of the target chip after the bonding wires according to the bonding wire path characteristics. ;

If there is no abnormality in the wire bonding path, extract the solder ball feature corresponding to the lead, and determine whether the solder ball of the target chip after the wire bonding is abnormal according to the solder ball feature;

If there is no abnormality in the solder balls of the target chip after the wire bonding, it is determined that there is no abnormality in the wire bonding of the target chip after the wire bonding.

Through the method provided by this embodiment, it can be determined whether the target chip after the wire bonding has abnormal wire bonding through the analysis of the number of wire bonding and the shape and size of the solder balls. Compared with manual detection of welding wires, it can ensure the comprehensiveness of welding wire detection and improve the accuracy of detection.

Optionally, the extracting the wire-bonding path characteristics of the leads, and determining whether there is an abnormality in the wire-bonding path of the target chip after the wire-bonding according to the wire-bonding path characteristics include:

According to the characteristics of the wire bonding path, determine whether the target chip after the wire bonding is disconnected;

If there is no broken wire, then extract the distance between every two leads in the wire path feature;

comparing said distance with a preset short-circuit distance;

If there is a wire distance less than the preset short-circuit distance, it is determined that the wire bonding path of the target chip after the wire bonding is abnormal;

If the distance between each two leads is greater than the preset short-circuit distance, it is determined that there is no abnormality in the bonding wire path of the target chip after the wire bonding.

Through the method provided by this embodiment, the distance between the leads can be confirmed, and compared with the preset short-circuit distance, it can be determined whether the target chip after the wire bonding may be abnormal. In order to avoid the appearance of short-circuited chips as far as possible, which will affect the experience of use.

Optionally, the extracting the features of the solder balls corresponding to the leads, and according to the features of the solder balls, determining whether there is an abnormality in the solder balls of the target chip after the wire bonding includes:

Determine the color of the solder ball according to the solder ball characteristics;

Determine whether there is oxidation in the solder ball according to the color of the solder ball;

If the solder balls are oxidized, it is determined that the solder balls of the target chip after the wire bonding are abnormal;

If there is no oxidation in the solder ball, then extract the solder ball diameter in the solder ball feature;

The diameter of the solder ball is matched with a preset diameter threshold, and if the diameter of the solder ball is lower than the preset diameter threshold, it is determined that there is no abnormality in the solder ball of the target chip after the wire bonding.

Through the method provided by this embodiment, when there is no abnormality in the above-mentioned lead wires and lead wire paths, the solder balls can be detected to avoid the situation that the solder balls are too large or too small to cause abnormal use of the target chip after wire bonding.

Optionally, according to the pulling force detection result, determining whether the wire bonding of the target chip after the wire bonding is qualified is firm, including:

Analyzing the tension detection result to determine the tightness of the leads of the target chip after the wire bonding;

Comparing the tightness of each lead wire with a preset tightness, to determine whether there is a lead wire smaller than the preset tightness;

If there are wires less than the preset tightness, it is determined that the wire bonding of the target chip after the wire bonding is not firm.

Through the method provided in this embodiment, through the detection of the push-pull machine, it is determined whether the target chip after the wire bonding is firmly bonded, which can further determine whether the target chip after the wire bonding is qualified, and reduce the detection error.

In a second aspect, the present application provides a chip bonding wire detection device, including:

An image receiving module, configured to receive the chip image uploaded by the image acquisition device;

A chip determination module, configured to determine whether the chip corresponding to the chip image is a target chip according to the chip image;

A wire bonding control module, configured to control the target chip to enter a wire bonding machine for wire bonding if the chip corresponding to the chip image is a target chip, to generate a target chip after wire bonding;

A machine starting module, configured to start a machine with a magnification effect to collect images of the target chip after the wire bonding;

The bonding wire analysis module is used to receive the image collected by the machine with the amplification effect, and determine whether there is abnormal bonding wire in the target chip after the bonding wire according to the image;

The push-pull machine control module is used to start the push-pull machine for pulling force detection if there is no abnormal wire bonding in the target chip after the wire bonding;

The tension analysis module is configured to receive the tension detection result of the push-pull machine, and determine whether the wire bonding of the target chip after the wire bonding is firm according to the tension detection result;

The qualification analysis module is configured to determine that the target chip after the wire bonding is qualified if the wire bonding of the target chip after the wire bonding is firm.

Optionally, the chip determination module is specifically used for:

Determine the model of the chip corresponding to the chip image through the chip image;

Determine a preset standard image according to the model; the preset standard image is an image corresponding to the model without any problem;

Matching the chip image with the preset standard image to determine the matching degree;

The matching degree is compared with a preset matching threshold, and if the matching degree is higher than the preset matching threshold, the chip corresponding to the chip image is used as a target chip.

Optionally, the chip determination module is also specifically used for:

performing grid division on the chip image and the preset standard image according to a preset method;

Comparing the screen color of each grid of the chip image with the screen color of each grid of the preset standard image to determine whether there are grids with inconsistent colors;

If there are grids with inconsistent colors, the color matching degree is determined according to the number of grids with inconsistent colors.

Optionally, the welding line analysis module is specifically used for:

Identify the leads of the image, and determine whether the number of bonding wires of the target chip after the bonding wires is correct;

If the number of bonding wires of the target chip after the wire bonding is correct, extract the bonding wire path characteristics of the leads, and determine whether there is an abnormality in the bonding wire path of the target chip after the bonding wires according to the bonding wire path characteristics. ;

If there is no abnormality in the wire bonding path, extract the solder ball feature corresponding to the lead, and determine whether the solder ball of the target chip after the wire bonding is abnormal according to the solder ball feature;

If there is no abnormality in the solder balls of the target chip after the wire bonding, it is determined that there is no abnormality in the wire bonding of the target chip after the wire bonding.

Optionally, the welding line analysis module is also specifically used for:

According to the characteristics of the wire bonding path, determine whether the target chip after the wire bonding is disconnected;

If there is no broken wire, then extract the distance between every two leads in the wire path feature;

comparing said distance with a preset short-circuit distance;

If there is a wire distance less than the preset short-circuit distance, it is determined that the wire bonding path of the target chip after the wire bonding is abnormal;

If the distance between each two leads is greater than the preset short-circuit distance, it is determined that there is no abnormality in the bonding wire path of the target chip after the wire bonding.

Optionally, the welding line analysis module is also specifically used for:

Determine the color of the solder ball according to the solder ball characteristics;

Determine whether there is oxidation in the solder ball according to the color of the solder ball;

If the solder balls are oxidized, it is determined that the solder balls of the target chip after the wire bonding are abnormal;

If there is no oxidation in the solder ball, then extract the solder ball diameter in the solder ball feature;

The diameter of the solder ball is matched with a preset diameter threshold, and if the diameter of the solder ball is lower than the preset diameter threshold, it is determined that there is no abnormality in the solder ball of the target chip after the wire bonding.

Optionally, the tension analysis module is specifically used for:

Analyzing the tension detection result to determine the tightness of the leads of the target chip after the wire bonding;

Comparing the tightness of each lead wire with a preset tightness, to determine whether there is a lead wire smaller than the preset tightness;

If there are wires less than the preset tightness, it is determined that the wire bonding of the target chip after the wire bonding is not firm.

In a third aspect, the present application provides an electronic device, including: a memory and a processor, where a computer program that can be loaded by the processor and execute the method of the first aspect is stored in the memory.

In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program capable of being loaded by a processor and executing the method of the first aspect.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present application. Those skilled in the art can also obtain other drawings based on these drawings without any creative effort.

FIG. 1 is a schematic diagram of an application scenario provided by an embodiment of the present application;

FIG. 2 is a flow chart of a chip bonding wire detection method provided by an embodiment of the present application;

FIG. 3 is a specific implementation flow of a chip bonding wire detection method provided by an embodiment of the present application;

FIG. 4 is a schematic structural diagram of a chip bonding wire detection device provided by an embodiment of the present application;

FIG. 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Apparently, the described embodiments are some of the embodiments of the present application, but not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of this application.

In addition, the term "and/or" in this article is only an association relationship describing associated objects, which means that there may be three relationships, for example, A and/or B may mean: A exists alone, A and B exist at the same time, There are three cases of B alone. In addition, the character "/" in this article, unless otherwise specified, generally indicates that the contextual objects are an "or" relationship.

The embodiments of the present application will be further described in detail below in conjunction with the accompanying drawings.

At present, IC detection is mostly carried out manually, but traditional manual detection requires a lot of manpower and material resources to train professional inspectors, and the controllability of manual detection is poor. In addition, long-term work will cause visual Fatigue, the accuracy of IC detection can not be controlled.

Based on this, the present application provides a chip bonding wire detection method, device, electronic equipment and storage medium.

Receive the chip image uploaded by the image acquisition device, determine the target chip according to the chip image, and then control the target chip to enter the wire bonding machine for wire bonding. The operation data of the wire bonding machine is received, and it is determined whether there is a target chip after wire bonding according to the operation data. If so, start a machine with an amplification effect for wire bonding inspection, and determine whether the target chip after the wire bonding is qualified according to the wire bonding inspection result. If qualified, then start the push-pull machine to carry out the tensile test, and according to the result of the tensile test, determine whether the lead wire welding of the target chip after the qualified bonding wire is firm, if it is firm, it means that the target chip bonding line is qualified and can be retained.

FIG. 1 is a schematic diagram of an application scenario provided by the present application. After the wire-bonding of the semiconductor chip is packaged, the wire-bonded chip can be inspected through the solution provided by the present application to determine whether the chip-bonding wire is qualified. In the application scene of this application, a pre-welding detection device and a post-welding detection device can be respectively set up for comprehensive detection of the chip. Whether the chip is defective. The post-soldering detection device includes a machine with amplifying effect and a push-pull machine. The machine with a magnifying effect can detect each lead wire of the chip after wire bonding to determine whether there is a missing wire or wrong welding; the push-pull machine can detect The firmness of the negative lead, and then determine whether the chip bonding wire is qualified. When it is determined by the image acquisition device that the chip has no defects, the wire bonding machine will be controlled to bond the chip. After the wire bonding, the chip first enters the machine with amplifying effect to detect the lead wires. Do a second test.

For specific implementation manners, reference may be made to the following embodiments.

FIG. 2 is a flow chart of a chip bonding wire detection method provided by an embodiment of the present application. The method of this embodiment can be applied to the server in the above scenario. As shown in Figure 2, the method includes:

S201. Receive the chip image uploaded by the image acquisition device.

The image acquisition device can be a device integrated with the wire bonding machine, which is set on the wire bonding machine. Before the wire bonding machine wires, the image acquisition device will first collect the image of the chip, and upload it to the server after the acquisition is completed.

In some implementation manners, the image acquisition device may also be separately arranged in front of the wire bonding machine, so as to ensure that the image of the chip is acquired in advance before the wire bonding machine wires bonding.

In some other implementation manners, the image acquisition device can be started all the time, and the characteristics of the chip are preset, and the corresponding picture will be uploaded only after the image acquisition device captures the characteristics related to the chip.

S202. According to the chip image, determine whether the chip corresponding to the chip image is a target chip.

The target chip can be understood as a chip that needs to be bonded but has not yet been bonded. In addition, the target chip can be a chip without any defect.

After receiving the chip image, the chip image can be identified to determine whether there is a chip in the uploaded picture, and if there is a chip, the chip can be inspected for defects. If there are defects, you can determine whether these defects affect the bonding wires of the chip.

In some implementations, a deep learning model can be established. In addition, a large number of defective chip images and corresponding chip defect types are collected, and these defective chip images and corresponding chip defect types are input as samples to the deep learning model. In the model, the deep learning model is trained so that the deep learning model can recognize the chip image and determine the type of defect. If a chip has no defects, the output content is empty, that is, no defect type will be output.

After the above-mentioned deep learning model is used to determine the type of defect in a certain chip, the range of the chip affected by all defects is identified according to the chip image. When the affected range exceeds a certain area, it can be considered that these defects seriously affect the use of the chip. Obsolete, do not perform wire welding. For example, if the defect coverage exceeds 25% of the total area of the chip, the chip can be scrapped. If the defect coverage accounts for less than 25% of the total chip area or if there are no defects, it can be used as a target chip.

S203. Control the target chip to enter the wire bonding machine for wire bonding, and generate the target chip after wire bonding.

After the target chip is determined through the above steps, the wire bonding machine can be started, and at the same time, the target chip can be transferred to the wire bonding machine for wire bonding to generate the wire bonded target chip.

S204, start the machine with zoom effect to collect the image of the target chip after wire bonding.

The machine with magnification effect can be an image acquisition device with magnification effect, which can enlarge the target chip after wire bonding, so that the state of each lead wire can be seen more clearly, so as to perform image acquisition on it.

S205. Receive an image collected by a machine with a magnification effect, and determine whether the target chip after wire bonding has abnormal wire bonding according to the image.

The image collected by the above-mentioned machine with magnification effect will also be uploaded to the server. Therefore, after receiving the image collected by the machine with magnification effect, the image can be identified to determine whether there is a lead wire leakage or a wrong wire of the target chip. In the case of soldering, for example, a certain pair of pins is not connected with a lead wire, or the lead wire of a certain pin is wrongly welded to another pin, and the two pins are not a pair, it can be considered that the welding wire is abnormal. In addition, it is also possible to determine whether there is a disconnection of the lead wire through image recognition, and if so, it can also be considered that there is an abnormality in the welding wire. In addition, it is also possible to determine whether the solder ball is standard through the image. If a certain solder ball is too large or too small, it can be considered that there is an abnormal solder wire.

S206 , if there is no abnormality in the wire bonding of the target chip where the wire bonding is completed, start the push-pull force machine to perform tension detection.

The push-pull machine can be used to detect the firmness of the chip's lead wire pull. If it is determined by analyzing the above image that there is no abnormality in the bonding wire, the push-pull machine can be started, and the target chip after the wire bonding is transferred to the push-pull machine for tension testing.

S207. Receive the tension detection result of the push-pull machine, and determine whether the wire bonding of the target chip after wire bonding is firm according to the tension detection result.

The pulling force detection result of the push-pull machine can be transmitted to the server. After receiving the pulling force detection result, the pulling force of the target chip after the wire bonding can be determined. If the pulling force reaches the standard, it can be considered that the wire bonding The target chip is solid, if the pulling force is smaller than this standard, it can be considered as not firm.

In some implementations, the tensile standard of each chip can be distinguished according to the model of the chip. Different types of chips may require different tensile forces to ensure the normal use of the chip. Therefore, the corresponding tensile standard can be determined by using the model of the chip, and then According to the result of the tension test, it is determined whether the target chip after the wire bonding is firm.

In other implementations, if the pulling force is greater than the corresponding pulling force standard of this model, the target chip after the wire bonding may be hit by the push-pull force machine after the pulling force test is completed, resulting in deformation or damage of the target chip after the wire bonding . In order to reduce the situation that the tension is strong but the chip is damaged due to excessive tension, after the tension detection lead wire is welded firmly, the image of the target chip after the wire bonding can be re-acquired through a machine with a magnification effect to analyze whether the wire bonding The state of the target chip.

S208. If the wire bonding of the target chip after wire bonding is firm, determine that the target chip after wire bonding is qualified.

Through the solution provided by this application, before the target chip is bonded, the chip is detected by the image acquisition device to determine the target chip, and the defective chips are minimized to be bonded, resulting in waste of resources. In addition, after welding, use a machine with amplifying effect to determine whether there is a problem with the welding wire. If there is no problem, use a push-pull machine to detect the pulling force of the welding wire. On the basis that there is no problem with the bonding wire, ensure that there is no problem with the tension of the bonding wire, thereby ensuring that the chip bonding wire is qualified and reducing the occurrence of defective products. For a specific implementation process, refer to FIG. 3 .

In some embodiments, the model of the chip corresponding to the chip image can be determined through the chip image; the preset standard image is determined according to the model; the chip image is matched with the preset standard image to determine the matching degree; the matching degree is compared with the preset The matching threshold is compared, and if the matching degree is higher than the preset matching threshold, the chip corresponding to the chip image is used as the target chip.

Since the number corresponding to the model of the chip can generally be engraved on a fixed position in the chip, after receiving the image of the chip through the above-mentioned image acquisition device, the image of the chip can be recognized to determine the character string existing in the image, so that according to the character string , to determine the corresponding model.

The preset standard image can be regarded as a chip image without any defects corresponding to each model. These preset standard images can be stored in the database, and at the same time have a one-to-one correspondence with the character strings of the corresponding models. The preset matching threshold can be considered as the minimum matching degree with the preset standard image but can be regarded as the matching degree corresponding to the target chip.

Specifically, after the chip image is determined, the model corresponding to the chip is determined according to the image recognition, so that the preset standard image corresponding to the character string is retrieved from the database, and then the chip image is matched with the preset standard image , it is determined that there is a difference, if there is a difference from the preset standard image, it can be considered that the chip is defective.

When the chip has a defect, the matching degree can be determined according to the area of the defect, and the matching degree is compared with the preset matching threshold. When the matching degree of the chip is higher than the preset matching threshold, it can be used as the target chip .

Through the method provided in this embodiment, when determining the target chip, the model of the chip can be determined first through image recognition, thereby calling the corresponding preset standard image, and then comparing to determine the matching degree. And set a preset matching threshold corresponding to each model, and when the matching degree is higher than the preset matching threshold, this chip can be used as the target chip. Determining the target chip through image recognition can reduce false detection problems that may be caused by human identification, reduce cost losses, and improve chip utilization.

In some embodiments, the chip image and the preset standard image can be divided into grids according to a preset method; Compare the screen colors of a grid to determine whether there are grids with inconsistent colors; if there are grids with inconsistent colors, determine the color matching degree according to the number of grids with inconsistent colors.

The grid division in this embodiment can be regarded as dividing the chip image into equal parts, which can ensure that the grid size of the chip is the same, and at the same time can ensure that the chip image corresponds to each grid of the preset standard image one-to-one. If the image features in a certain grid of the chip image are inconsistent with the image features of the corresponding grid of the preset standard image, it can be considered that this grid does not match the preset standard image.

In some implementations, the image features are inconsistent, and the chips may have different colors, there may be oxidation, and the chips may be missing. When performing image matching, it is not necessary to consider whether the inconsistency is the color inconsistency or the missing chip. As long as the inconsistency is inconsistent, it can be considered as a mismatch.

In other implementations, several image features corresponding to oxidation and some image features corresponding to chip loss can be set in advance, and these image features are input into a new deep learning model, and the new deep learning model is trained so that This deep learning model can determine whether the defect of the chip is caused by oxidation or missing chip by recognizing the chip image. Therefore, the chip image was input into this new deep learning model after meshing the chip image, so as to determine the reason for the inconsistency of the mesh inconsistent with the preset standard image. When comparing the matching degree with the preset matching threshold, the preset matching threshold can be set differently, for example, as a preset color matching threshold and a preset missing matching threshold. Wherein, if the matching degree brought by the grid inconsistent with the preset standard image color is higher than the preset color matching threshold, it can be considered that this chip can be used as the target chip. If the matching degree brought about by the missing chip exceeds the preset missing matching threshold, it can also be considered that this chip can be used as the target chip.

Through the method provided by this embodiment, the chip image can be identified and distinguished by the color of the image, so as to determine the difference between the chip and the standard chip, and then determine whether this chip can be used as a target chip. Through this image recognition method, the matching degree of the chip can be determined as comprehensively as possible, and the error caused by human detection can be reduced.

In some embodiments, the wires of the image can be identified to determine whether the number of wires of the target chip after the wire bonding is correct; Wire path features to determine whether there is an abnormality in the wire bonding path of the target chip after wire bonding; if there is no abnormality in the wire bonding path, extract the solder ball feature corresponding to the lead, and determine the solder ball feature of the target chip after wire bonding according to the solder ball feature. Whether there is an abnormality in the ball; if there is no abnormality in the solder ball of the target chip after the wire bonding, it is determined that there is no abnormal wire bonding in the target chip after the wire bonding.

The lead wire can be understood as the wire used when bonding the chip. Because the pins of each type of chip that need to be bonded are known, after the type of the chip is determined, the pins of the chip can be determined according to the type of the chip. Thereby determine the number of leads required according to the number of pins. If the number of leads does not correspond, no matter whether the number is large or small, it can be considered that the number of bonding wires is incorrect, and the target chip after this bonding wire can be considered unqualified.

If the number of welding wires is standard and there is no under-welding or over-welding, other aspects can be detected, such as whether the welding wires are connected in the wrong position. Because the pins generally exist in pairs, in the process of welding wires, the leads of the general welding wires are in one-to-one correspondence, or the way of welding wires is preset in advance, so in the case of a standard number of welding wires , the wire path feature corresponding to the lead can be extracted, and the wire path feature can include features such as the length and direction of the wire. In order to conveniently determine whether the characteristics of the welding wire path are qualified, the characteristics of the welding wire path of the preset standard image can be extracted and compared one by one to determine whether there is an abnormality in the welding wire path.

During the wire bonding process, each pin will produce a solder ball due to the wire bonding. The size of this solder ball also has certain regulations during the wire bonding process. If it is too large or too small, it is not qualified. In addition, the shape of the solder ball There will also be certain restrictions. Generally, if the shape of the solder ball is a standard sphere, it can be considered qualified. In order to determine whether the shape of the solder ball is qualified, the shape of the solder ball in the target chip after wire bonding can be compared with the shape of the solder wire corresponding to the preset standard chip to determine whether the solder ball is abnormal.

Therefore, when it is determined that there is no abnormality in the wire path through the above method, the solder ball features of the solder balls corresponding to each lead can be extracted, and the size and shape of the solder balls can be determined by identifying and analyzing the solder ball features. Compare the size of the precipitated solder ball with the size of the solder ball feature corresponding to the preset standard image, and determine whether the solder ball size of the target chip after wire bonding is consistent with the size of the solder ball corresponding to the preset standard image, if the size is consistent , it can be considered qualified. If not, it can be considered unqualified.

In the case of the same size, the shape of the solder ball can be identified to determine whether the target chip after wire bonding is consistent with the solder ball shape corresponding to the preset standard image. If they are consistent, it can be considered qualified.

Through the method provided by this embodiment, it can be determined whether the target chip after the wire bonding has abnormal wire bonding through the analysis of the number of wire bonding and the shape and size of the solder balls. Compared with manual detection of welding wires, it can ensure the comprehensiveness of welding wire detection and improve the accuracy of detection.

In some embodiments, it is possible to determine whether there is a broken wire in the target chip after the wire bonding according to the characteristics of the wire bonding path; Distance; compare the distance with the preset short-circuit distance; if there is a lead distance less than the preset short-circuit distance, it is determined that the bonding wire path of the target chip after the wire bonding is abnormal; the distance between every two leads is greater than the preset If the short-circuit distance is short, it is determined that there is no abnormality in the bonding wire path of the target chip after the wire bonding.

The preset short-circuit distance can be considered as the distance between the two leads caused by the deviation of the welding wire or the excessive length of the welding. At this time, when the bonding wire is biased or the bonding wire is too long, the distance between the two leads is too short, which may cause a bonding short circuit in the subsequent use of the chip.

When it is determined that there is no abnormality in the number of leads, and there is a one-to-one correspondence between the leads, and there is no path abnormality, it can be determined whether each lead is disconnected according to the characteristics of the welding wire path. If so, it can be explained that there is a disconnection , this lead is invalid at this time. The distance between leads can be extracted from the wire path feature if there are no lead breaks. Compare the distance between each two leads to the preset short distance. If the distance between every two leads is less than the preset short-circuit distance, a bonding short-circuit may occur, and at this time, it can be considered that there is an abnormality in the bonding wire path. However, if the distance between the leads is greater than the preset short-circuit distance, it can be considered that the chip will not have a bonding short circuit during subsequent use, and it can be considered that there is no abnormality in the bonding wire path.

Through the method provided by this embodiment, the distance between the leads can be confirmed, and compared with the preset short-circuit distance, it can be determined whether the target chip after the wire bonding may be abnormal. In order to avoid the appearance of short-circuited chips as far as possible, which will affect the experience of use.

In some embodiments, the color of the solder ball can be determined according to the characteristics of the solder ball; according to the color of the solder ball, it is determined whether there is oxidation in the solder ball; if there is oxidation, it is determined that there is an abnormality in the solder ball of the target chip after the wire bonding; if the copper wire If there is no oxidation, then extract the solder ball diameter in the solder ball feature; match the solder ball diameter with the preset diameter threshold, if the solder ball diameter is above the preset diameter threshold, then determine that the solder ball of the target chip after the wire bonding is not There is an exception.

The preset diameter threshold can be regarded as the normal diameter of the solder ball of the target chip after the wire bonding, wherein the preset diameter threshold can be the maximum value of the solder ball or the minimum value of the solder ball. When it is greater than this maximum value or less than this minimum value, it indicates that there is an abnormality in the solder ball. The maximum or minimum value corresponding to the preset diameter threshold can be determined by the diameter of the lead wire, for example, the maximum value can be 4 times the diameter of the lead wire, and the minimum value can be 2 times the diameter of the lead wire.

If the solder ball is oxidized, the surface of the solder ball may be dull or dark, and the solder ball may also appear red. Therefore, the color of each solder ball can be determined through the extracted features of the solder ball. If the above-mentioned change occurs in the color of the solder ball, it can indicate that the solder ball is oxidized. If there is no above-mentioned change in the color of the solder ball, it can be explained that there is no oxidation of the solder ball.

When it is determined that there is no oxidation of the solder ball, the diameter of the solder ball can be determined according to the characteristics of the solder ball. In addition, the diameter of the lead corresponding to the solder ball is determined through the characteristics corresponding to the lead, so as to obtain the preset diameter threshold. If the preset diameter threshold is the maximum value of the solder ball, then the diameter of the solder ball is compared with the preset diameter threshold, and if the diameter of the solder ball is greater than the preset diameter threshold, it can be determined that the size of the solder ball is abnormal. In addition, if the preset diameter threshold is the minimum value of the solder ball, then the diameter of the solder ball is compared with the preset diameter threshold, and if the diameter of the solder ball is smaller than the preset diameter threshold, it can be determined that the size of the solder ball is abnormal. On the contrary, it means that there is no abnormality.

Through the method provided by this embodiment, when there is no abnormality in the above-mentioned lead wires and lead wire paths, the solder balls can be detected to avoid the situation that the solder balls are too large or too small to cause abnormal use of the target chip after wire bonding.

In some embodiments, the tension detection result can be analyzed to determine the tightness of the leads of the target chip after the qualified wire bonding; the tightness of each lead is compared with the preset tightness to determine whether there is a wire less than the preset tightness of the lead; if there is a lead less than the preset tightness, it is determined that the bonding wire of the target chip after the wire bonding is not firm.

The preset tightness can be considered as the tightness corresponding to the firmness of the target chip bonding wire after the wire bonding. If it is greater than this tightness, it may cause the chip to be hit and deformed during the detection process. If it is smaller than this tightness, it may cause The target chip after the wire bonding is easily damaged in subsequent use.

Specifically, after obtaining the tension detection result uploaded by the push-pull machine, the tension detection result is analyzed to determine the degree of tightness corresponding to the tension detection result. Then the tightness of each lead of the target chip after wire bonding is compared with the preset tightness. When the tightness of the leads is less than the preset tightness, the target chip after the wire bonding may be easily damaged in subsequent use. Indicates that the welding wire is not strong.

If the tightness of the lead wire is greater than the preset tightness, the machine with the zoom effect can be controlled to reacquire the image of the target chip after the wire bonding, and it is determined whether the target chip after the wire bonding is detected by the push-pull machine. Leave marks of being hit. If there is, it can also indicate that the welding wire is abnormal and cannot be used normally.

Through the method provided in this embodiment, through the detection of the push-pull machine, it is determined whether the target chip after the wire bonding is firmly bonded, which can further determine whether the target chip after the wire bonding is qualified, and reduce the detection error.

FIG. 4 is a schematic structural diagram of a chip bonding wire detection device provided by an embodiment of the present application. As shown in FIG. Welding wire control module 403 , machine startup module 404 , welding wire analysis module 405 , push-pull force machine control module 406 , pulling force analysis module 407 , and qualification analysis module 408 .

An image receiving module 401, configured to receive the chip image uploaded by the image acquisition device;

A chip determining module 402, configured to determine whether the chip corresponding to the chip image is a target chip according to the chip image;

The wire bonding control module 403 is configured to, if the chip corresponding to the chip image is a target chip, control the target chip to enter a wire bonding machine for wire bonding, and generate a target chip after wire bonding;

A machine starting module 404, configured to start a machine with a magnification effect to collect images of the target chip after the wire bonding;

The wire bonding analysis module 405 is configured to receive the image collected by the machine with the amplification effect, and determine whether there is an abnormal wire bonding in the target chip after the wire bonding according to the image;

The push-pull machine control module 406 is used to start the push-pull machine for pulling force detection if the target chip after the wire-bonding does not have abnormal wire-bonding;

The tension analysis module 407 is configured to receive the tension detection result of the push-pull machine, and determine whether the wire bonding of the target chip after the wire bonding is firm according to the tension detection result;

The qualification analysis module 408 is configured to determine that the target chip after the wire bonding is qualified if the wire bonding of the target chip after the wire bonding is firm.

Optionally, the chip determination module 402 is specifically configured to:

Determine the model of the chip corresponding to the chip image through the chip image;

Determine a preset standard image according to the model; the preset standard image is an image corresponding to the model without any problem;

Matching the chip image with the preset standard image to determine the matching degree;

The matching degree is compared with a preset matching threshold, and if the matching degree is higher than the preset matching threshold, the chip corresponding to the chip image is used as a target chip.

Optionally, the chip determination module 402 is specifically further configured to:

performing grid division on the chip image and the preset standard image according to a preset method;

Comparing the screen color of each grid of the chip image with the screen color of each grid of the preset standard image to determine whether there are grids with inconsistent colors;

If there are grids with inconsistent colors, the color matching degree is determined according to the number of grids with inconsistent colors.

Optionally, the welding line analysis module 405 is specifically used for:

Identify the leads of the image, and determine whether the number of bonding wires of the target chip after the bonding wires is correct;

If the number of bonding wires of the target chip after the wire bonding is correct, extract the bonding wire path characteristics of the leads, and determine whether there is an abnormality in the bonding wire path of the target chip after the bonding wires according to the bonding wire path characteristics. ;

If there is no abnormality in the wire bonding path, extract the solder ball feature corresponding to the lead, and determine whether the solder ball of the target chip after the wire bonding is abnormal according to the solder ball feature;

If there is no abnormality in the solder balls of the target chip after the wire bonding, it is determined that there is no abnormality in the wire bonding of the target chip after the wire bonding.

Optionally, the welding line analysis module 405 is also specifically used for:

According to the characteristics of the wire bonding path, determine whether the target chip after the wire bonding is disconnected;

If there is no broken wire, then extract the distance between every two leads in the wire path feature;

comparing said distance with a preset short-circuit distance;

If there is a wire distance less than the preset short-circuit distance, it is determined that the wire bonding path of the target chip after the wire bonding is abnormal;

If the distance between each two leads is greater than the preset short-circuit distance, it is determined that there is no abnormality in the bonding wire path of the target chip after the wire bonding.

Optionally, the welding line analysis module 405 is also specifically used for:

Determine the color of the solder ball according to the solder ball characteristics;

Determine whether there is oxidation in the solder ball according to the color of the solder ball;

If the solder balls are oxidized, it is determined that the solder balls of the target chip after the wire bonding are abnormal;

If there is no oxidation in the solder ball, then extract the solder ball diameter in the solder ball feature;

The diameter of the solder ball is matched with a preset diameter threshold, and if the diameter of the solder ball is lower than the preset diameter threshold, it is determined that there is no abnormality in the solder ball of the target chip after the wire bonding.

Optionally, the tension analysis module 407 is specifically used for:

Analyzing the tension detection result to determine the tightness of the leads of the target chip after the wire bonding;

Comparing the tightness of each lead wire with a preset tightness, to determine whether there is a lead wire smaller than the preset tightness;

If there are wires less than the preset tightness, it is determined that the wire bonding of the target chip after the wire bonding is not firm.

FIG. 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application. As shown in FIG. 5 , the electronic device 500 of this embodiment may include: a memory 501 and a processor 502 .

The memory 501 stores computer programs that can be loaded by the processor 502 and execute the methods in the foregoing embodiments.

Wherein, the processor 502 is connected to the memory 501, such as through a bus.

Optionally, the electronic device 500 may further include a transceiver. It should be noted that, in practical applications, the transceiver is not limited to one, and the structure of the electronic device 500 does not limit the embodiment of the present application.

The processor 502 may be a CPU (Central Processing Unit, central processing unit), a general-purpose processor, a DSP (Digital Signal Processor, a data signal processor), an ASIC (Application Specific Integrated Circuit, an application specific integrated circuit), an FPGA (Field Programmable Gate Array, field programmable gate array) or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various illustrative logical blocks, modules and circuits described in connection with the present disclosure. The processor 502 may also be a combination that implements computing functions, for example, a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

A bus may include a path for transferring information between the above-mentioned components. The bus may be a PCI (Peripheral Component Interconnect, Peripheral Component Interconnect Standard) bus or an EISA (Extended Industry Standard Architecture, Extended Industry Standard Architecture) bus or the like. The bus can be divided into address bus, data bus, control bus and so on. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

The memory 501 can be ROM (Read Only Memory, read-only memory) or other types of static storage devices that can store static information and instructions, RAM (Random Access Memory, random access memory) or other types of static storage devices that can store information and instructions Dynamic storage devices can also be EEPROM (Electrically Erasable Programmable Read Only Memory, Electrically Erasable Programmable Read-Only Memory), CD-ROM (Compact DiscRead Only Memory, CD-ROM) or other CD storage, CD storage (including compact CD, laser discs, compact discs, digital versatile discs, blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures that can be accessed by a computer media, but not limited thereto.

The memory 501 is used to store the application program code for executing the solution of the present application, and the execution is controlled by the processor 502 . The processor 502 is configured to execute the application program codes stored in the memory 501, so as to implement the contents shown in the foregoing method embodiments.

Among them, electronic devices include but are not limited to: mobile phones, notebook computers, digital broadcast receivers, PDA (personal digital assistant), PAD (tablet computer), PMP (portable multimedia player), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc. Mobile terminals such as digital TVs, desktop computers, etc. and fixed terminals. Also for servers etc. The electronic device shown in FIG. 5 is only an example, and should not limit the functions and scope of use of this embodiment of the present application.

The electronic device in this embodiment can be used to execute the method in any one of the above embodiments, and its implementation principle and technical effect are similar, and will not be repeated here.

The present application also provides a computer-readable storage medium storing a computer program capable of being loaded by a processor and executing the methods in the above embodiments.

Those of ordinary skill in the art can understand that all or part of the steps for implementing the above method embodiments can be completed by program instructions and related hardware. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above-mentioned method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other various media that can store program codes.

Claims (10)

1. A chip bonding wire detection method is characterized in that it is applied to a chip bonding wire detection system, and the chip bonding wire detection system includes a server, an image acquisition device, a machine with amplifying effect, a push-pull machine and a wire bonding machine, The chip bonding wire detection method is executed by a server, and the method includes: receiving the chip image uploaded by the image acquisition device; According to the chip image, determine whether the chip corresponding to the chip image is a target chip; If the chip corresponding to the chip image is a target chip, then controlling the target chip to enter the wire bonding machine for wire bonding to generate a target chip after wire bonding; Start the machine with the amplification effect to collect the image of the target chip after the wire bonding; Receiving the image collected by the machine with the magnification effect, and determining whether the target chip after the wire bonding has abnormal wire bonding according to the image; If there is no abnormal wire bonding in the target chip after the wire bonding, then start the push-pull machine to perform tension detection; receiving the pulling force detection result of the push-pull machine, and determining whether the wire bonding of the target chip after the wire bonding is firm according to the pulling force detection result; If the wire bonding of the target chip after wire bonding is firm, it is determined that the target chip after wire bonding is qualified. 2. The method according to claim 1, wherein the determining whether the chip corresponding to the chip image is a target chip according to the chip image comprises: Determine the model of the chip corresponding to the chip image through the chip image; Determine a preset standard image according to the model; the preset standard image is an image corresponding to the model without any problem; Matching the chip image with the preset standard image to determine the matching degree; The matching degree is compared with a preset matching threshold, and if the matching degree is higher than the preset matching threshold, the chip corresponding to the chip image is used as a target chip. 3. The method according to claim 2, wherein the matching degree comprises a color matching degree, and the matching of the chip image and the preset standard image to determine the matching degree comprises: performing grid division on the chip image and the preset standard image according to a preset method; Comparing the screen color of each grid of the chip image with the screen color of each grid of the preset standard image to determine whether there are grids with inconsistent colors; If there are grids with inconsistent colors, the color matching degree is determined according to the number of grids with inconsistent colors. 4. The method according to claim 1, wherein, according to the image, determining whether the target chip after the wire bonding has abnormal wire bonding comprises: Identify the leads of the image, and determine whether the number of bonding wires of the target chip after the bonding wires is correct; If the number of bonding wires of the target chip after the wire bonding is correct, extract the bonding wire path characteristics of the leads, and determine whether there is an abnormality in the bonding wire path of the target chip after the bonding wires according to the bonding wire path characteristics. ; If there is no abnormality in the wire bonding path, extract the solder ball feature corresponding to the lead, and determine whether the solder ball of the target chip after the wire bonding is abnormal according to the solder ball feature; If there is no abnormality in the solder balls of the target chip after the wire bonding, it is determined that there is no abnormality in the wire bonding of the target chip after the wire bonding. 5. The method according to claim 4, wherein the extraction of the bonding wire path feature of the lead wire determines whether the bonding wire path of the target chip after the bonding wire exists according to the bonding wire path feature exceptions, including: According to the characteristics of the wire bonding path, determine whether the target chip after the wire bonding is disconnected; If there is no broken wire, then extract the distance between every two leads in the wire path feature; comparing said distance with a preset short-circuit distance; If there is a wire distance less than the preset short-circuit distance, it is determined that the wire bonding path of the target chip after the wire bonding is abnormal; If the distance between each two leads is greater than the preset short-circuit distance, it is determined that there is no abnormality in the bonding wire path of the target chip after the wire bonding. 6. The method according to claim 4, wherein the extracting the solder ball feature corresponding to the lead wire determines whether there is an abnormality in the solder ball of the target chip after the wire bonding according to the solder ball feature, include: Determine the color of the solder ball according to the solder ball characteristics; Determine whether there is oxidation in the solder ball according to the color of the solder ball; If the solder balls are oxidized, it is determined that the solder balls of the target chip after the wire bonding are abnormal; If there is no oxidation in the solder ball, then extract the solder ball diameter in the solder ball feature; The diameter of the solder ball is matched with a preset diameter threshold, and if the diameter of the solder ball is lower than the preset diameter threshold, it is determined that there is no abnormality in the solder ball of the target chip after the wire bonding. 7. The method according to claim 1, wherein, according to the pulling force detection result, determining whether the wire bonding of the target chip after the wire bonding is firm includes: Analyzing the tension detection result to determine the tightness of the leads of the target chip after the wire bonding; Comparing the tightness of each lead wire with a preset tightness, to determine whether there is a lead wire smaller than the preset tightness; If there are wires less than the preset tightness, it is determined that the wire bonding of the target chip after the wire bonding is not firm. 8. A chip bonding wire detection device, characterized in that it comprises: An image receiving module, configured to receive the chip image uploaded by the image acquisition device; A chip determination module, configured to determine whether the chip corresponding to the chip image is a target chip according to the chip image; A wire bonding control module, configured to control the target chip to enter a wire bonding machine for wire bonding if the chip corresponding to the chip image is a target chip, to generate a target chip after wire bonding; A machine starting module, configured to start a machine with a magnification effect to collect images of the target chip after the wire bonding; The bonding wire analysis module is used to receive the image collected by the machine with the amplification effect, and determine whether there is a bonding wire abnormality in the target chip after the bonding wire according to the image; The push-pull machine control module is used to start the push-pull machine for pulling force detection if there is no abnormal wire bonding in the target chip after the wire bonding; The tension analysis module is configured to receive the tension detection result of the push-pull machine, and determine whether the wire bonding of the target chip after the wire bonding is firm according to the tension detection result; The qualification analysis module is configured to determine that the target chip after the wire bonding is qualified if the wire bonding of the target chip after the wire bonding is firm. 9. An electronic device, comprising: a memory and a processor; The memory is used to store program instructions; The processor is configured to call and execute the program instructions in the memory, and execute the chip bonding wire detection method according to any one of claims 1-7. 10. A computer-readable storage medium, characterized in that, a computer program is stored in the computer-readable storage medium; when the computer program is executed by a processor, the method according to any one of claims 1-7 is realized. A chip bonding wire detection method.