Etching compensation method, etching compensation system and deep learning system

Abstract

The invention provides an etching compensation method, an etching compensation system and a deep learning system, and the etching compensation method comprises the steps: shooting a workpiece which is subjected to etching processing, so as to obtain a workpiece image; measuring the workpiece according to the workpiece image to obtain line measurement information of the workpiece; generating etching compensation information according to the line measurement information; and transmitting the etching compensation information to an exposure device or an etching device. According to the invention, after etching treatment, full-edition and comprehensive measurement of the workpiece can be carried out, and then through the functions of full-automatic rapid compensation value setting, full-automatic production intermediate monitoring and instant feedback adjustment, the effects of rapidly entering mass production, rapidly reaching optimal compensation value setting, monitoring in intelligent production and improving and stabilizing quality are achieved.

Description

Technical Field

[0001] The present invention relates to a compensation method, a compensation system, and a deep learning system, and particularly to an etching compensation method, an etching compensation system, and a deep learning system for an etching process. Background Art

[0002] In modern manufacturing processes, as electronic products become increasingly miniaturized and integrated, the frequency of circuit signals is increasing day by day, and the circuits of circuit boards and integrated circuits are becoming increasingly thinner. Due to the increasing thinness of the circuits, the consistency of the cross-sectional area of the circuits is becoming increasingly crucial for circuit characteristics such as resistance and impedance. Slight changes will result in the final electrical performance not meeting expectations.

[0003] During the circuit etching process, factors such as the original designed circuit layout density and circuit shape will cause differences in the flow and contact conditions of the etching solution, resulting in different etching rates. Therefore, appropriate adjustments need to be made for the circuit characteristics of different regions presented after etching.

[0004] Traditionally, production personnel make initial settings for circuit etching based on experience and adjust parameters through actual measurement of etching results. The traditional method is limited by measurement efficiency and production pressure, resulting in insufficient measurement positions and imperfect parameter verification, leading to circuit production quality not meeting expectations. Summary of the Invention

[0005] The main objective of the present invention is to provide an etching compensation method, including: photographing a workpiece that has undergone an etching process to obtain a workpiece image; measuring the workpiece based on the workpiece image to obtain line measurement information of the workpiece; generating etching compensation information based on the line measurement information; and transmitting the etching compensation information to an exposure device or an etching device.

[0006] Another objective of the present invention is to provide an etching compensation system, including a measurement device and a compensation device. The measurement device photographs a workpiece that has undergone an etching process to obtain a workpiece image and analyzes the workpiece image to generate line measurement information of the workpiece. The compensation device is coupled to the measurement device and generates etching compensation information based on the line measurement information.

[0007] Another objective of the present invention is to provide a deep learning system for cooperating with the above-described etching compensation system. The deep learning system includes a data storage device and a processing device. The data storage device is connected to the compensation device and receives the etching compensation information to establish a sample database. The processing device is connected to the data storage device to access the sample database. The processing device includes a deep neural network, and the deep neural network is trained using the etching compensation information stored in the sample database.

[0008] Therefore, the present invention can perform full-panel and comprehensive measurement of the workpiece after the etching process, and then achieve the effects of quickly entering mass production, quickly achieving the best compensation value setting, monitoring during intelligent production, and improving and stabilizing the quality through the functions of automatic and rapid compensation value setting, automatic production intermediate monitoring, and immediate feedback adjustment. In addition, the etching compensation method of the present invention has a wide range of applications. In addition to performing compensation for the circuit design image, it can also adjust the etching process parameters, allowing the operator to select the compensation method according to actual needs. Among them, for the compensation of the circuit design image, it can be implemented on an exposure machine that requires an exposure negative, or it can be implemented on an exposure machine that does not require an exposure negative. Furthermore, the etching compensation system of the present invention can be trained through a deep learning system, and then perform measurement and etching compensation operations with artificial intelligence, reducing labor costs and achieving automated operations. Brief Description of the Drawings

[0009] Figure 1 It is a block diagram of the etching compensation system in the first embodiment of the present invention.

[0010] Figure 2 It is a schematic cross-sectional view of the circuit in the present invention.

[0011] Figure 3 It is an isometric view of the circuit in the present invention.

[0012] Figure 4 It is a schematic flowchart of the etching compensation method in the first embodiment of the present invention.

[0013] Figure 5 It is a combined block diagram of the deep learning system and the etching compensation system in the second embodiment of the present invention.

[0014] Figure 6 It is a schematic flowchart of the training process of the deep learning system in the present invention.

[0015] Figure 7 It is a block diagram of the etching compensation system in the third embodiment of the present invention.

[0016] Figure 8 It is a schematic flowchart of the etching compensation method in the third embodiment of the present invention.

[0017] The labels in the figures are as follows:

[0018] 10, 10' Etching compensation system

[0019] 1 Measuring device

[0020] 11 Image capturing device

[0021] 12 Image processing device

[0022] 2 Compensation device

[0023] 3 Image generation device

[0024] 4 Exposure device

[0025] 5 Image detection device

[0026] DB Image database

[0027] W Workpiece

[0028] UW Upper width of the line

[0029] LW Lower width of the line

[0030] TH Thickness of the line

[0031] L Length of the line

[0032] M Deep learning system

[0033] M1 Data storage device

[0034] M2 Processing device

[0035] 6 Developing device

[0036] 7 Etching device

[0037] OLI Original exposure image

[0038] CC Circuit compensation image

[0039] WI Workpiece image

[0040] MK Mask Detailed implementation mode

[0041] A detailed description and technical content of the present invention will be described below in conjunction with the accompanying drawings. Furthermore, for the convenience of illustration, the scale of the drawings in the present invention is not necessarily drawn according to the actual scale and there is an exaggerated situation. The drawings and their scales are not intended to limit the scope of the present invention.

[0042] The present invention can be used in the process of workpiece circuit etching. When the circuit etching is completed, the compensation value of the morphology after circuit etching is confirmed through automatic optical inspection, so as to perform corresponding corrections immediately and comprehensively during the processing process to ensure that the produced workpieces can be standardized. On the other hand, it can also avoid errors caused by environmental changes in a large number of processes. The workpieces in the present invention can include, for example, but are not limited to, printed circuit boards (PCBs), flexible printed circuit boards (FPCs), ceramic substrates, integrated circuit wafers or integrated circuit chips. The variations of the embodiments are not within the scope intended to be limited by the present invention.

[0043] The following is an explanation of the first embodiment of the present invention. Please first refer toFigure 1 , which is a block diagram of the etching compensation system of the present invention, as shown in the figure.

[0044] This embodiment provides an etching compensation system 10 for measuring various data of the circuit on a workpiece W after etching, such as information including line width, line pitch, line length, shape, circuit direction, circuit thickness, cross-sectional area, circuit volume, or circuit type, etc. In the first embodiment, the etching compensation system 10 includes a measuring device 1, a compensation device 2, an image generating device 3, an exposure device 4, an image detecting device 5, and an image database DB.

[0045] The measuring device 1 captures the workpiece W after etching treatment to obtain a workpiece image, and analyzes the workpiece image to generate the circuit measurement information of the workpiece W. In the first embodiment, the measuring device 1 includes an image capturing device 11 and an image processing device 12 connected or coupled to the image capturing device 11.

[0046] In the first embodiment, the image capturing device 11 may include, for example, but is not limited to, a line scan camera or an area scan camera. In another embodiment, the image capturing device 11 of the measuring device 1 may include a dual-lens device (for example, including an overhead lens and a side view lens), which captures and obtains the upper width and the lower width of the circuit in the overhead direction, obtains the circuit height through capturing in the side view direction, and calculates and obtains the circuit cross-sectional area based on the above information. In another embodiment, the measuring device 1 may include a three-dimensional image camera, and generates circuit measurement information through 3D scanning technology (such as Time of Flight (TOF), Triangulation, Stereo Vision). The manner of obtaining the circuit measurement information is not within the scope of the present invention.

[0047] In the first embodiment, the image processing device 12 is connected to the image capturing device 11 to obtain the workpiece image of the workpiece W and generate the circuit measurement information of the workpiece W. In the first embodiment, the circuit measurement information includes line width, line pitch, circuit direction, circuit thickness, circuit volume, circuit type, or other such information. In the first embodiment, the image processing device 12 of the measuring device 1 can measure the workpiece W across the entire board according to the workpiece image to obtain the circuit measurement information of all the circuits on the workpiece W.

[0048] In the first embodiment, the compensation device 2 is coupled to the measurement device 1, and an etching compensation information is calculated and generated based on the line measurement information. Specifically, the compensation device 2 calculates the etching compensation information according to the difference between the line measurement information and a preset line information. The preset line information is, for example, the line information of a motherboard or a good product. The etching compensation information is, for example, the compensation value that each mask on the exposure image should be corrected at the line position where the difference is located. The conversion of the compensation value (from the difference to the compensation value) is performed by substituting the correction parameter according to the etching environment. The correction parameter can be, for example, machine parameters (such as LDI accuracy, incident angle, output power, etc.), reticle registration error, material characteristics and uniformity, environmental parameters (such as environmental temperature, humidity, etc.), or any other conditions that may affect the final etching result. After substituting all or part of the variation parameters, a compensation value is generated.

[0049] In the first embodiment, the image generation device 3 is coupled to the compensation device 2, generates a line compensation image based on the etching compensation information, and provides it to the exposure device 4 to update the exposure negative for the etching process. Specifically, the image generation device 3 can obtain the original exposure image pre-stored in the image database DB, or obtain the original exposure image from an external device. The original exposure image referred to here is the exposure negative used when performing the corresponding etching process. Thus, based on the original exposure image as the standard image, the original exposure image is compensated according to the corresponding etching compensation information, and the updated line compensation image is output. For example, in the line compensation image, when the compensation value at the coordinate (x, y) is +2μm and the original mask is 10μm, the mask corresponding to the coordinate (x, y) is modified to 12μm for compensation; in practice, the compensation value can also be divided into two blocks on the left and right sides of the corresponding coordinate to separately perform compensation correction on the two side boundaries of the mask, which is not limited in the present invention.

[0050] In the first embodiment, the exposure device 4 receives the line compensation image from the image generation device 3, updates the exposure negative in the image database DB with the received line compensation image, and performs exposure and development based on the updated exposure negative. For example, it can include, but is not limited to, an inner layer exposure machine, an outer layer exposure machine, a solder mask exposure machine, a parallel light exposure machine, or a non-parallel light exposure machine.

[0051] It should be noted that in other embodiments, after the image generation device 3 generates the line compensation image, it can be directly provided to an exposure device 4 that does not require an exposure negative, such as a digital direct imaging exposure machine (using digital exposure technology, Digital Lithography Technology, DLT), a laser direct imaging device (Laser Direct Imaging Device, LDI), etc., for exposure and development, which is not limited in the present invention.

[0052] It should be noted that, in the first embodiment, the line compensation image can be one of a Computer Aided Manufacturing (CAM) file, a GERBER file, an EXCELLON file, an ODB++ file, an IPC-2581 file, a DXF file, a BOM file, a Pick and Place file, and a Netlist file, or a combination of these files, which is not limited in the present invention.

[0053] In the first embodiment, the etching compensation system 10 of the present invention can be integrated on an automatic optical inspection device, and compensation correction can be performed immediately according to the finished product while performing automatic optical inspection, thereby immediately correcting the etching process. In the first embodiment, the image detection device 5 is connected or coupled to the measurement device 1, and the workpiece W is detected for defects according to the workpiece image to obtain the defect detection information of the workpiece W. In the first embodiment, defect detection can be performed by comparing the workpiece image with the master image (or good product image) in the image database DB via a traditional algorithm (such as image subtraction) to confirm the position of the defect, or by inputting the workpiece W into a trained neural network (such as a deep learning model or a machine learning model) for defect identification. The variations of the embodiments are not within the scope intended to be limited by the present invention. In other embodiments, the etching compensation system 10 can also be implemented as an independent inspection device, for example, arranged at the front end or the rear end of the automatic optical inspection device. The variations of the embodiments are not within the scope intended to be limited by the present invention. In other embodiments, the etching compensation system 10 may not include the image detection device 5 and only perform image measurement and etching compensation operations on the workpiece W, which is not limited in the present invention.

[0054] In the first embodiment, the image processing device 12, the compensation device 2, and the image generation device 3 of the measurement device 1 can be implemented by different computers, servers, PLCs, or similar independent devices. In other embodiments, the image processing device 12, the compensation device 2, and the image generation device 3 can be co-constructed as a single computer, server, PLC, or similar independent device, and after the processor loads the storage unit, it executes the program stored in the storage unit. The processor can be, for example, a Central Processing Unit (CPU), or other programmable general-purpose or special-purpose microprocessors, Digital Signal Processors (DSPs), programmable controllers, Application Specific Integrated Circuits (ASICs), Programmable Logic Devices (PLDs), or other similar devices, or a combination of these devices, which is not limited in the present invention.

[0055] In other embodiments, after measuring the workpiece W using an automatic optical detection device and obtaining the line measurement information, the line measurement information can be provided to an external computer (not shown in the figure) to generate a line compensation image, and then provided to the exposure device 4 or cause the external computer to generate an updated exposure negative for providing to the exposure device 4. Specifically, for example, after the system integrates and analyzes the results of training, a corresponding database (such as a look-up table) is created and stored in the external computer. When the external computer receives the line measurement information, it finds the corresponding data (such as a compensation value or a target value) from the database based on the detected value in the line measurement information, and generates a line compensation image based on the data.

[0056] Regarding the process executed by the compensation system of the present invention, please refer to Figure 2 and Figure 3 , which are the schematic diagrams of the line profile and the line isometric view in the full-panel measurement of the present invention, as shown in the figure. In the first embodiment, the image generation device 3, including but not limited to, can generate a line compensation image through the following compensation methods.

[0057] After the image processing device 12 obtains the workpiece image from the image capture device 11, it stores the workpiece image in the image database DB to prepare for measurement.

[0058] After the workpiece image enters the scheduling, the measuring device 1 generates line measurement information of the workpiece W from the workpiece image. The line measurement information is information about each line in the workpiece image, including line width, line pitch, line direction, line thickness, line volume, or line type, etc. During full-panel measurement, the lines can be identified by searching for the color or shape of the lines in the workpiece image to segment the boundary between the lines and the substrate on the workpiece image. The method of this image segmentation is not limited within the scope of the present invention. The "full-panel measurement" described in the present invention refers to image measurement of the entire workpiece. Specifically, it can be to use a line-scan camera to scan the workpiece and measure the entire workpiece image after stitching all the line segment images together, or it can be to form the workpiece image of the full panel by taking partial regions of the workpiece in batches through a surface-scan camera, or to take the full-panel image of the workpiece at one time, which is not limited in the present invention.

[0059] Specifically, the image processing device 12 obtains the information of the lines through the following method. The image processing device 12 can first perform an image pre-processing program (such as image enhancement, noise removal, contrast enhancement, edge enhancement, feature capture, image compression, image conversion, etc.), and segment or capture the boundary of the image after the image pre-processing program to divide the region of interest (ROI). The method of capturing the region of interest can be, for example, binarization or through neural networks such as machine learning systems or deep learning systems. After being trained by the system, the line regions and substrate regions in the workpiece image are segmented. It should be noted that the workpiece image mentioned here is not necessarily a single image, but can also be two or more images, and multiple workpiece images are used to analyze the distribution information of the lines in three-dimensional space for subsequent analysis.

[0060] In the first embodiment, the cross-sectional area of the line can be obtained by calculating the line width (such as including the upper width UW and the lower width LW of the line, etc.) and the line thickness TH, for example, using the trapezoidal area formula. Further, in the first embodiment, the line volume can be obtained by calculating the line cross-sectional area and the line length L. In other embodiments, for a more accurate calculation method, the line volume can be obtained by calculating the cross-sectional area of each sampled line and the length of each sampled line.

[0061] In the first embodiment, when the image processing device 12 performs measurement, pixel coordinates can be used as an index to record the values of each circuit on the workpiece image. Considering the computing efficiency of the image processing device 12, at least a single pixel can be used as the minimum unit for recording the circuit size value. In the case where the computing efficiency of the image processing device 12 is limited, the circuit size value of the corresponding section can also be recorded in sections with a distance of two or more pixels. Such an implementation method is not within the scope of the present invention. In other embodiments, the circuit measurement information can also be recorded using custom-defined coordinates as an index, which is not limited in the present invention.

[0062] After the image processing device 12 completes all measurements, the compensation device 2 compares the measured values with the circuit values of a standard chip (such as a master chip) pre-stored in the image database DB to determine the difference between the workpiece image and the standard chip. Specifically, before performing the above measurements, the required standard chip can be sent to the device of the present invention to first perform full-panel measurement through the measurement device 1, so as to obtain various circuit measurement information of the standard chip first, and store the circuit measurement information in the image database DB as the expected value of each circuit coordinate with coordinates as an index. When formally performing compensation measurement, the compensation device 2 subtracts the measured value of the workpiece image captured from the expected value at the corresponding position of the standard chip to obtain the difference indexed by coordinates. In the first embodiment, the standard chip can be a master chip or a good chip, which is not limited in the present invention.

[0063] Furthermore, when the compensation device 2 obtains the difference between the measured value of each circuit coordinate and the expected value, it can calculate the compensation value at the corresponding coordinate position in the exposure image according to the difference. The compensation value specifically refers to the compensation value required to correct the circuit size to the expected value for the circuit mask (such as the masked or unmasked part) on the exposure image. The correction parameters for converting the difference to the compensation value can be machine parameters (such as LDI accuracy, incident angle, output power, etc.), mask registration error, material characteristics and uniformity, environmental parameters (such as environmental temperature, humidity, etc.) or any other conditions that may affect the final etching result. After substituting all or part of the variable parameters, the compensation value is generated.

[0064] After obtaining the compensation values for the entire plate, the image generation device 3 corrects the circuit mask (such as widening or reducing the width of the circuit mask) at the mapping coordinate positions corresponding to the compensation values based on the original exposure image (specifically, to generate the original exposure negative for the workpiece circuit). When all circuit masks are corrected, a circuit compensation image is generated and sent to the exposure device 4 to update the exposure negative for the etching process.

[0065] Please refer to Figure 4, which is a schematic flowchart of the etching compensation method in the first embodiment of the present invention. The etching compensation method includes the following execution steps.

[0066] First, photograph the workpiece after etching treatment to obtain a workpiece image (step S01); in the first embodiment, the workpiece includes substrates such as printed circuit boards, flexible circuit boards, high-density interconnect technology circuit boards, touch panels, or integrated circuit wafers, etc., and the embodiments are not within the scope of the present invention to be limited.

[0067] Next, measure the workpiece according to the workpiece image to obtain a line measurement information of the workpiece (step S02); in the first embodiment, further including the situation of the image detection device, while the image processing device measures the workpiece image, it can simultaneously perform defect detection on the workpiece image to obtain a defect detection information of the workpiece.

[0068] Next, generate an etching compensation information according to the line measurement information (step S03); in the first embodiment, the etching compensation information includes the coordinate values of the lines and the compensation values corresponding to the positions of the coordinate values.

[0069] Finally, transmit the etching compensation information to an exposure device (step S04). The exposure device includes but is not limited to a traditional exposure machine that requires an exposure negative or a direct imaging exposure machine that does not require an exposure negative.

[0070] The present invention discloses a deep learning system in the second embodiment. Please refer to Figure 5 and Figure 6 , which are the combined block diagram of the deep learning system and the etching compensation system of the present invention and the training flowchart of the deep learning system, as shown in the figure.

[0071] This embodiment discloses a deep learning system M for cooperating with the etching compensation system 10 described above. In the second embodiment, the deep learning system M includes a data storage device M1 and a processing device M2 connected to the data storage device M1.

[0072] The data storage device M1 is connected to the etching compensation system 10, and the workpiece image and the line compensation image are provided by the etching compensation system 10 to establish a sample database. Specifically, each time the etching compensation system 10 performs etching treatment correction, the corresponding values of the workpiece image corresponding to the part number are transmitted to the data storage device M1 and stored in the sample database. Among them, the corresponding values may include etching compensation information such as the workpiece image, the line measurement information (including coordinate positions) of the full-board line size of the workpiece, the full-board compensation value (including coordinate positions) of the exposure negative, and the updated line compensation image.

[0073] The processing device M2 accesses the sample database of the data storage device M1. The processing device M2 includes a deep neural network, and the deep neural network is trained using the workpiece images and circuit compensation images stored in the sample database. In other embodiments, the deep neural network can be trained using the workpiece images and etching compensation information stored in the sample database.

[0074] The deep neural network in the present invention can be, for example, but not limited to, LeNet model, AlexNet model, GoogleNet model, or VGG model (Visual Geometry Group), etc., which is not limited in the present invention. Specifically, the deep neural network can be trained to convert a workpiece image into a circuit compensation image. The workpiece image is used as the input of the deep neural network, and the corresponding circuit compensation image is output according to the workpiece image. The deep neural network of the present invention is used to input the workpiece image into the deep neural network for forward propagation to obtain an excitation response, and to calculate the difference between the excitation response result and the circuit compensation image after circuit correction, so as to obtain the response error therein and perform backpropagation to update the weights. Through the repeated iteration of the above two steps, until the response of the deep neural network to the input reaches the target range.

[0075] In the second embodiment, the training process of the deep learning system in the present invention is as follows: First, the workpiece W is transferred to the exposure device 4, and the exposure device 4 performs an exposure process on the workpiece W according to the photomask MK generated from the original exposure image OLI, so as to transfer the photomask pattern onto the workpiece W. Subsequently, the workpiece W with the generated photomask pattern is transferred to a developing device 6 and an etching device 7. First, the developing device 6 develops the workpiece W to generate an etching window, and then the etching device 7 etches the workpiece W. After etching, the workpiece W is transferred to the etching compensation system 10, and the workpiece image WI is obtained through the etching compensation system 10, and full-panel measurement is performed according to the workpiece image WI to obtain circuit measurement information. Based on the result of the full-panel measurement, the etching compensation system 10 generates etching compensation information according to the circuit measurement information, and generates the circuit compensation image CC to be imaged according to the etching compensation information. For example, in the second embodiment, the original exposure image OLI can be adjusted according to the etching compensation information to generate the circuit compensation image CC to be imaged. Subsequently, the circuit compensation image CC is transferred to the exposure device 4 to update the exposure negative / image of the exposure device 4.

[0076] On the other hand, in the second embodiment, the etching compensation system 10 can transfer the captured workpiece image WI and the corresponding circuit compensation image CC of the workpiece image WI to the deep learning system M for training.

[0077] As a result of the above training, the trained deep neural network will acquire the skill of converting the workpiece image into a circuit compensation image, so as to obtain an artificial intelligence model for workpiece etching measurement.

[0078] Please refer to Figure 7 and Figure 8 , which are respectively a block diagram of the etching compensation system and a flowchart of the etching compensation method in the third embodiment of the present invention.

[0079] In the third embodiment, the etching compensation system 10' is substantially the same as the etching compensation system 10 in the first embodiment, except that: in step S03, after the compensation device 2 generates the etching compensation information; in step S04, an etching process correction parameter is generated according to the etching compensation information; and then in step S05, the etching process correction parameter is transmitted to the etching device 7, so as to update the process parameters when the etching device 7 performs the etching process.

[0080] In the third embodiment, the etching process correction parameter includes one or any combination of the type of etching solution, the concentration of etching solution, the volume of etching solution, the pressure or direction of the nozzle for spraying the etching solution, the speed of stirring the etching solution, the gas composition, the gas pressure, the radio frequency power, the exhaust rate, the temperature and humidity of the etching process, and the execution time of the etching process, which is not limited in the present invention.

[0081] For example, the problem of insufficient etching of the workpiece W can be improved by increasing the concentration and / or flow rate of the etching solution, increasing the contact time between the workpiece W and the etching solution, etc. On the contrary, the problem of over-etching can be improved by reducing the concentration and / or flow rate of the etching solution, shortening the contact time between the workpiece W and the etching solution, etc. The actual application of the above process correction parameters is not limited in the present invention, and is hereby stated in advance.

[0082] To sum up, the present invention can perform full-panel and comprehensive measurement of the workpiece after the etching process, and then through the functions of automatic and rapid compensation value setting, automatic production intermediate monitoring, and instant feedback adjustment, achieve the effects of quickly entering mass production, quickly reaching the optimal compensation value setting, monitoring in intelligent production, and improving and stabilizing the quality. In addition, the etching compensation method of the present invention has a wide range of applications. In addition to performing circuit design image compensation, it can also adjust the etching process parameters, allowing the operator to select the compensation method according to actual needs. For the compensation of the circuit design image, it can be implemented on an exposure machine that requires an exposure negative or an exposure machine that does not require an exposure negative. Furthermore, the etching compensation system of the present invention can be trained by a deep learning system, and then perform measurement and etching compensation operations with artificial intelligence, reducing labor costs and achieving automated operations.

[0083] The present invention has been described in detail above. As mentioned above, the above is only a preferred embodiment of the present invention, and it cannot be used to limit the scope of implementation of the present invention. That is, all equivalent changes and modifications made according to the scope of the patent application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. An etching compensation system, characterized in that: Include: A measuring device photographs a workpiece after an etching process to obtain a workpiece image, and analyzes the workpiece image to generate a line measurement information of the workpiece; as well as A compensation device is coupled to the measurement device and generates etching compensation information according to the circuit measurement information.

2. The etching compensation system according to claim 1, characterized in that: The measuring device measures the workpiece on the entire board according to the workpiece image to obtain the circuit measurement information of the workpiece, and the etching compensation system also includes: an image detection device coupled to the measuring device, which detects defects on the workpiece according to the workpiece image to obtain defect detection information of the workpiece.

3. The etching compensation system according to claim 1, characterized in that: It also includes: an image generating device, which generates a line compensation image according to the etching compensation information, and the line compensation image includes one of a CAM file, a GERBER file, an EXCELLON file, an ODB++ file, an IPC-2581 file, a DXF file, a BOM file, a Pick an Place file and a Netlist file, or any combination thereof.

4. The etching compensation system according to claim 1, characterized in that: The invention also comprises: an image generating device, wherein the image generating device generates a line compensation image according to the etching compensation information, and the line compensation image is transmitted to an exposure device to update the exposure film of the etching process.

5. The etching compensation system according to claim 1, characterized in that: The compensation device generates an etching process correction parameter according to the etching compensation information, and the etching process correction parameter is transmitted to an etching device to update the process parameters of the etching process, and the etching process correction parameter includes one of the type of etching liquid, the concentration of the etching liquid, the volume of the etching liquid, the pressure or direction of the nozzle spraying the etching liquid, the speed of stirring the etching liquid, the gas composition, the gas pressure, the radio frequency power, the exhaust rate, the temperature and humidity of the etching process, and the execution time of the etching process, or any combination thereof.

6. The etching compensation system according to claim 1, characterized in that: Also includes: an image generating device, the image generating device generating a line compensation image according to the etching compensation information; and An exposure device receives the line compensation image from the image generating device, wherein the exposure device includes a digital direct imaging exposure machine, a laser direct imaging device, an inner layer exposure machine, an outer layer exposure machine, a solder mask exposure machine, a parallel light exposure machine or a non-parallel light exposure machine.

7. The etching compensation system according to claim 1, characterized in that: The compensation device calculates the etching compensation information according to the difference between the circuit measurement information and a preset circuit information, wherein the circuit measurement information includes line width, line spacing, circuit direction, circuit thickness, circuit volume or circuit type.

8. A deep learning system, characterized in that: To cooperate with the etching compensation system as described in claims 1 to 7, the deep learning system comprises: a data storage device connected to the compensation device, receiving the etching compensation information to establish a sample database; and A processing device is connected to the data storage device to access the sample database, wherein the processing device includes a deep neural network, and the deep neural network is trained using the etching compensation information stored in the sample database.

9. A deep learning system, characterized in that: To cooperate with the etching compensation system as claimed in claim 4, the deep learning system comprises: a data storage device, receiving the line compensation image to establish a sample database; and A processing device is connected to the data storage device to access the sample database, wherein the processing device includes a deep neural network, and the deep neural network is trained using the workpiece image and the line compensation image stored in the sample database.

10. An etching compensation method, characterized in that: include: photographing a workpiece after an etching process to obtain a workpiece image; Measuring the workpiece according to the workpiece image to obtain line measurement information of the workpiece; Generate etching compensation information according to the circuit measurement information; as well as The etching compensation information is transmitted to an exposure device or an etching device.

11. The etching compensation method according to claim 10, characterized in that: The step of measuring the workpiece according to the workpiece image includes: measuring the workpiece on the entire board surface according to the workpiece image to obtain the circuit measurement information of the workpiece; And the etching compensation method further comprises: defect detecting the workpiece according to the workpiece image to obtain defect detection information of the workpiece.

12. The etching compensation method according to claim 10, characterized in that: It also includes: generating a line compensation image according to the etching compensation information, wherein the line compensation image includes one of a CAM file, a GERBER file, an EXCELLON file, an ODB++ file, an IPC-2581 file, a DXF file, a BOM file, a Pick an Place file and a Netlist file or any combination thereof.

13. The etching compensation method according to claim 10, characterized in that: The method also includes: generating a line compensation image according to the etching compensation information, and transmitting the line compensation image to the exposure device to update the exposure film of the etching process.

14. The etching compensation method according to claim 10, characterized in that: It also includes: generating an etching process correction parameter according to the etching compensation information, and the etching process correction parameter is transmitted to the etching device to update the process parameters of the etching process, and the etching process correction parameter includes one of the type of etching liquid, the concentration of the etching liquid, the volume of the etching liquid, the pressure or direction of the nozzle spraying the etching liquid, the speed of stirring the etching liquid, the gas composition, the gas pressure, the radio frequency power, the exhaust rate, the temperature and humidity of the etching process, and the execution time of the etching process, or any combination thereof.

15. The etching compensation method according to claim 10, characterized in that: The exposure device includes a digital direct imaging exposure machine, a laser direct imaging device, an inner layer exposure machine, an outer layer exposure machine, a solder mask exposure machine, a parallel light exposure machine or a non-parallel light exposure machine.

16. The etching compensation method according to claim 10, characterized in that: The step of generating the etching compensation information according to the line measurement information includes: calculating the etching compensation information according to the difference between the line measurement information and a preset line information, wherein the line measurement information includes line width, line spacing, line direction, line thickness, line volume or line type.

Images