A method for ensuring cleanliness of chip plastic packaging

Abstract

The present invention discloses a method for ensuring the cleanliness of chip plastic packaging, which specifically relates to the field of chip packaging, including obtaining empirical data related to machine models, plastic packaging material models and processes; obtaining a mold surface image during the plastic packaging process; preprocessing the image, extracting key features of the image, comparing it with the clean mold surface image, and estimating the cleanliness of the mold surface; and requiring or forcing the operator to wash the mold in combination with historical empirical data. The present invention uses an image sensor to collect images of multiple mold surfaces, and compares them with the image of the clean mold surface to determine whether to wash the mold; it not only uses the empirical parameters obtained in historical production, but also ensures that there is not much modification to the production line, and can quickly adjust to enter the production state; uses the image to quickly estimate the cleanliness of the mold surface, and provides a decision basis for whether to wash the film, thereby ensuring the cleanliness of the plastic packaging mold, thereby improving the yield rate of the plastic packaging process of chip packaging, and improving production efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of chip packaging, and more specifically, to a method for ensuring the cleanliness of chip plastic packaging. Background Art

[0002] In the chip packaging process, plastic encapsulation is the process of placing the lead frame after lead bonding on the molding machine under pressure and high temperature, using epoxy resin as a plastic encapsulation material for sealing, so that the chip is protected from light, moisture, deformation, heat and radiation during use, which will eventually lead to the degradation of electronic device performance. Plastic encapsulation first places the lead frame between the upper and lower molds; puts the plastic encapsulation material into the glue inlet; tightens the upper and lower molds and applies a certain amount of pressure to form a cavity; the injection rod squeezes the high-temperature melted plastic encapsulation material into the cavity; after a certain period of time, the plastic encapsulation material solidifies and completes the plastic encapsulation.

[0003] After multiple plastic sealing, a certain amount of plastic sealing material will remain on the surface of the mold. Due to the thermosetting nature of the material, it cannot be molded again. Therefore, the residual plastic sealing material will cause defects and deformations in the subsequent plastic sealing, which directly indicate that the product is unqualified. Less residual plastic sealing material will also reduce the bonding force between the new plastic sealing material and the mold surface, causing plastic sealing stratification and bubbles, affecting heat dissipation efficiency, and ultimately affecting product quality and service life. In addition, after multiple plastic sealing, the residual plastic sealing material will also affect the service life of the mold, causing the mold to deform prematurely under the huge plastic sealing pressure, affecting the plastic sealing effect. Therefore, the cleanliness of the mold surface needs to be ensured during the plastic sealing process.

[0004] The common practice now is to set a film washing time and mold washing frequency based on experience, and stipulate that the laminating machine should wash the mold every certain time and after every certain number of batches. If it is set completely according to experience, the plastic sealing quality of the first few plastic sealing batches before mold washing will be poor due to a small amount of plastic sealing material, which will affect the service life of the product. However, film washing is a time-consuming operation, and it is necessary to balance production efficiency and film washing frequency.

[0005] In actual production environments, chip laminators from different manufacturers are used. Some models are older and have limited computing power, so the algorithm needs to be simple and effective enough. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method for ensuring the cleanliness of chip plastic packaging, which can be used on equipment of different manufacturers and different models, facing different types of plastic packaging materials. After a batch of packaging is completed, an image sensor is used to detect the cleanliness of the mold surface. When the surface cleanliness deteriorates to a certain extent, the machine is stopped and the operator is forced to start cleaning the mold.

[0007] To achieve the above object, the present invention provides the following technical solution: a method for ensuring the cleanliness of chip plastic packaging, the specific steps are as follows:

[0008] Step S1: obtaining the empirical film washing conditions set by the process and related to the machine model, the plastic packaging material model and the process;

[0009] Step S2: During the plastic sealing process, the mold surface image is acquired by an image sensor;

[0010] Step S3: extracting the image texture features, comparing the current mold surface image with the clean mold surface image, and estimating the cleanliness of the mold surface to be tested;

[0011] Step S4: Compare the mold washing conditions obtained in step S1 and suggest or force the operator to wash the mold.

[0012] In a preferred embodiment, in step S1, the film washing conditions related to the current equipment, molding materials and process are obtained from the database, specifically including the film washing interval time, the number of film washing interval molding times, the forced film washing cleanliness and the recommended film washing cleanliness.

[0013] In a preferred embodiment, in step S2, two image sensors are used to respectively acquire images of the upper and lower mold surfaces.

[0014] In a preferred embodiment, the specific steps of step S3 are as follows:

[0015] Step S3.1: Standardize and grayscale the current mold closing image and the clean mold closing image, the formula is:

[0016]

[0017] y gray =0.299y r +0.587y g +0.114y b Formula 2;

[0018] Step S3.2: Calculate the grayscale co-occurrence matrix of the normalized grayscale image, and extract the angular second moment ASM, inverse variance IDM, contrast Con and entropy ENT of the matrix as representative features. The formula is as follows:

[0019]

[0020]

[0021] Step S3.3: Compare the texture features p of the clean mold surface image and the image to be detected, and estimate the cleanliness s of the mold surface to be detected. The formula is as follows:

[0022]

[0023] In a preferred embodiment, in step S4, the film washing conditions obtained in step S1 are compared to determine whether mold washing is required; if forced mold washing is required, the machine production is stopped until the mold washing is completed; if mold washing is not required, the process is terminated; the specific implementation steps are as follows:

[0024] Step S41: Determine whether the cleanliness of the mold surface obtained in step S3 is lower than the cleanliness of the forced film washing. If so, jump to step S45, otherwise jump to step S42;

[0025] Step S42: Determine whether the mold surface cleanliness obtained in step S3 is lower than the recommended film washing cleanliness. If so, jump to step S49, otherwise jump to step S43;

[0026] Step S43: Determine whether the number of productions since the last film washing exceeds the set interval plastic sealing number; if so, jump to step S49, otherwise jump to step S44;

[0027] Step S44: determine whether the production time from the last membrane washing exceeds the set interval time; if so, jump to step S49, otherwise end;

[0028] Step S45: lock the machine, stop the machine operation, remind the operator to clean up, and then jump to step S46;

[0029] Step S46: Determine whether the mold cleaning is completed, if so, jump to step S48, otherwise jump to step S47;

[0030] Step S47: Lock the machine, stop the machine operation, wait for the cleaning to be completed, and then jump to step S46;

[0031] Step S48: Unlock the machine, and the machine starts production, and then ends;

[0032] Step S49: Remind the operator that the membrane needs to be washed, and then jump to step S43.

[0033] Technical effects and advantages of the present invention:

[0034] The present invention uses an image sensor to collect images of multiple mold surfaces, compares them with images of clean mold surfaces, and estimates the cleanliness of the mold surfaces to be detected, so as to determine whether the mold should be cleaned, whether it is a major clean or a minor clean. It not only uses the empirical parameters obtained from historical production, but also ensures that there is no major modification to the production line, and can quickly adjust to the production state. Although the cleanliness estimated by the algorithm of the present invention has a certain error, it is sufficient for the chip packaging plastic sealing process. Using the present invention, the overall film washing time can be reduced, the productivity and the packaging yield rate can be improved, and it is conducive to standardizing the production process and facilitating management. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 The present invention is a general implementation flow chart of a method for ensuring the cleanliness of chip plastic packaging.

[0036] Figure 2 The present invention is a schematic diagram of a machine platform used in a method for ensuring the cleanliness of chip plastic packaging.

[0037] Figure 3 The present invention is a flowchart for determining whether to recommend or even force the operator to clean the mold.

[0038] The reference numerals in the drawings are: 1 upper mold, 2 lower mold, 3 clamp, 4 push rod, 5 image sensor. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] The present invention provides a method for ensuring the cleanliness of chip plastic packaging, comprising the following steps:

[0041] Step S1: Obtain the machine model, material model, process and other empirical film washing conditions set by the process;

[0042] Step S2: acquiring upper and lower mold surface images through an image sensor;

[0043] Step S3: extracting the key features of the image, comparing the current mold surface image with the clean mold surface image, and estimating the cleanliness of the mold surface to be tested;

[0044] Step S4: Compare the film washing conditions obtained in step S1 to determine whether the mold needs to be washed; if the mold needs to be washed, stop the machine production until the mold is washed; if the mold does not need to be washed, end this process;

[0045] Specifically, the present invention uses two image sensors 5 to respectively acquire images of the surfaces of the upper mold 1 and the lower mold 2, wherein both sides of the upper mold 1 and the lower mold 2 are clamped by a clamp 3, and a push rod 4 is further provided at the bottom of the clamp 3 corresponding to the lower mold 2. Figure 2 As shown in the figure, the texture of the image will be greatly different due to the presence of the plastic packaging material, so the difference in image texture information is used to estimate the overall cleanliness of the mold surface;

[0046] Step S3.1: Standardize and grayscale the current mold closing image and the clean mold closing image, the formula is:

[0047]

[0048] y gray =0.299y r +0.587y g +0.114y b Formula 2;

[0049] Where x is the image to be standardized, mean is the mean of each channel of the image to be standardized, and std is the standard deviation of each channel of the image to be standardized; y gray is the grayscale value of the image after grayscale conversion, y r ,y g and b They are the standardized red, green and blue channel data respectively;

[0050] In order to ensure that high-quality images can be obtained under different camera parameters and different lighting conditions, and taking into account the limited computing power of old equipment, the fastest pixel normalization is used. However, the normalization method used in the present invention is not limited to this. The same histogram normalization, standard deviation normalization, etc. can all achieve the functions of the present invention.

[0051] Step S3.2, respectively calculate the gray level co-occurrence matrix of the two images and extract features; the gray level co-occurrence matrix measures the joint probability distribution of a pixel pair with a spatial relationship.

[0052] P(i,j|d,θ) represents the probability that the gray value starts at i and ends at j when the spatial distance d and the direction θ are given; the present invention selects 8 groups of gray-level co-occurrence matrices with spatial distances of 1 and 2 and directions of 0°, 45°, 90° and 135° respectively;

[0053] For each set of gray level co-occurrence matrices, the second-order moment ASM, inverse variance IDM, contrast Con and entropy ENT are extracted as representative features: the angular second-order moment is a measure of gray level distribution uniformity and texture coarseness, and is defined as follows:

[0054]

[0055] The inverse variance measures the degree of cleanliness and regularity of the texture. The clearer the texture, the stronger the regularity, and the easier it is to describe, the larger the inverse variance. Its definition is as follows:

[0056]

[0057] The entropy measures the randomness of the information contained in the image. The more complex the image, the greater the entropy. It is defined as follows:

[0058]

[0059] Contrast reflects the local changes in the image. The clearer the texture, the greater the contrast, and the greater the contrast. It is defined as follows:

[0060]

[0061] Step S3.3: Compare the texture features p of the clean mold surface image and the image to be detected, and estimate the cleanliness s of the mold surface to be detected;

[0062] Compare the 8 groups of features obtained in step S3.2, a total of 32 features, to estimate the cleanliness of the mold surface to be tested;

[0063]

[0064] Wherein T is a diagonal matrix, the sum of the squares of its diagonal elements is 1, which ensures that the range of the image texture features is not changed, and its function is to change the weight of the features used in step S3.2; the definition of the feature vector is as follows:

[0065]

[0066] Different molds of the same machine need to be cleaned at different times. A thorough cleaning takes 1 hour, which seriously affects production efficiency. Therefore, we have added a simple cleaning option, which takes 5 minutes and can significantly remove the plastic sealing material on the mold surface, called small cleaning, and the corresponding thorough cleaning is called large cleaning. When the cleanliness of a mold surface reaches the set forced film washing cleanliness, the machine must be stopped for large cleaning. When the cleanliness of only one mold surface reaches the recommended cleaning cleanliness but does not reach the forced cleaning cleanliness, small cleaning is recommended. When the cleanliness of multiple mold surfaces reaches the recommended cleaning cleanliness but does not reach the forced cleaning cleanliness, small cleaning can be performed, but large cleaning is recommended.

[0067] like Figure 3 In step S4, the film washing conditions obtained in step S1 are compared to determine whether mold washing is required; if forced mold washing is required, the machine production is stopped until the mold washing is completed; if mold washing is not required, the process is terminated; the specific implementation steps are as follows:

[0068] Step S41: determine whether the cleanliness of the mold surface obtained in step S3 is lower than the cleanliness of the forced film washing. If so, jump to step S45, otherwise jump to step S42;

[0069] Step S42: Determine whether the cleanliness of the mold surface obtained in step S3 is lower than the recommended cleanliness of film washing; if so, jump to step S49, otherwise jump to step S43;

[0070] Step S43: determine whether the number of productions since the last film washing exceeds the set interval plastic sealing number; if so, jump to step S49, otherwise jump to step S44;

[0071] Step S44: determine whether the production time from the last membrane washing exceeds the set interval time; if so, jump to step S49, otherwise end;

[0072] Step S45: lock the machine, stop the machine operation, remind the operator to clean the machine, and then jump to step S46; Step S46: determine whether the mold cleaning is completed, if so, jump to step S48, otherwise jump to step S47;

[0073] Step S47: lock the machine, stop the machine operation, wait for the cleaning to be completed, then jump to step S46; Step S48: unlock the machine, the machine starts production, and then ends;

[0074] Step S49: Remind the operator that the membrane needs to be washed, and then jump to step S43.

[0075] The present invention only selects four statistical features, namely, the angular second moment, inverse variance, contrast and entropy in the gray-level co-occurrence matrix, as the basis to estimate the cleanliness of the mold surface to be detected; there are old machine tools on the current production line, and the computing power is limited, so the present invention chooses to use image processing methods to estimate the cleanliness. Although the cleanliness estimated by this method is somewhat different from the actual surface cleanliness and cannot be used for measurement, it is sufficient to meet the needs of plastic packaging production in chip packaging. This example provides the most suitable implementation solution we know of, and the present invention can also be implemented by selecting other statistical features.

[0076] Finally, a few points should be explained: First, in the description of this application, it should be noted that, unless otherwise specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, which may refer to mechanical connection or electrical connection, or internal communication between two components, or direct connection. "upper", "lower", "left", "right", etc. are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship may change;

[0077] Secondly: In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0078] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for ensuring the cleanliness of chip plastic packaging, characterized in that: The specific steps are as follows: Step S1: Obtaining experience data of machine model, plastic packaging material model and process; Step S2: During the plastic sealing process, an image of the mold surface is obtained; Step S3: preprocess the image, extract key features of the image, compare it with the clean mold surface image, and estimate the cleanliness of the mold surface; Step S4: Step S41: Determine whether the cleanliness of the mold surface obtained in step S3 is lower than the cleanliness of the forced film washing. If so, jump to step S45, otherwise jump to step S42; Step S42: Determine whether the cleanliness of the mold surface obtained in step S3 is lower than the recommended cleanliness of film washing. If so, jump to step S49, otherwise jump to step S43; Step S43: Determine whether the number of productions since the last film washing exceeds the set interval plastic sealing number; if so, jump to step S49, otherwise jump to step S44; Step S44: determine whether the production time from the last membrane washing exceeds the set interval time; if so, jump to step S49, otherwise end; Step S45: lock the machine, stop the machine operation, remind the operator to clean it thoroughly, and then jump to step S46; Step S46: Determine whether the mold cleaning is completed, if so, jump to step S48, otherwise jump to step S47; Step S47: Lock the machine, stop the machine operation, wait for the complete cleaning to be completed, and then jump to step S46; Step S48: Unlock the machine, and the machine starts production, and then ends; Step S49: Remind the operator that the membrane needs to be washed, and then jump to step S43.

2. A method for ensuring the cleanliness of chip plastic packaging according to claim 1, characterized in that: The specific steps of step S3 are as follows: Step S3.1: Standardize and grayscale the current mold closing image and the clean mold closing image, the formula is: y gray =0.299y r +0.587y g +0.114y b ; Where x is the image to be standardized, mean is the mean of each channel of the image to be standardized, and std is the standard deviation of each channel of the image to be standardized; y gray is the grayscale value of the image after grayscale conversion, y r ,y g and b They are the standardized red, green and blue channel data respectively; Step S3.2: Calculate the grayscale co-occurrence matrix of the normalized grayscale image, and extract the angular second moment ASM, inverse variance IDM, contrast Con and entropy ENT of the matrix as representative features. The formula is as follows: P(i,jd,θ) represents the probability that the gray value starts at i and ends at j when the spatial distance d and the direction θ are given; Step S3.3: Compare the texture features p of the clean mold surface image and the image to be detected, and estimate the cleanliness s of the mold surface to be detected. The formula is as follows: T in step S3.3 is a diagonal matrix, the sum of the squares of its diagonal elements is 1, and its function is to change the weights of the features used in step S3.

2.

3. A method for ensuring the cleanliness of chip plastic packaging according to claim 2, characterized in that: The cleanliness of the mold surface can be estimated simply and quickly using the texture information of the image.

4. A method for ensuring the cleanliness of chip plastic packaging according to claim 1, characterized in that: The empirical data of step S1 specifically include the film washing interval time, the number of times of plastic sealing between film washing, the forced film washing cleanliness and the recommended film washing cleanliness.

Images