Kerf Recognition via Image Subtraction on Dicing Tape
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in accurately recognizing a kerf on a dicing tape due to interference from its adhesive pattern, which complicates image processing and alignment during the hairline alignment process in cutting machines.
Innovation Solution
A recognition method involving bonding a workpiece to a larger dicing tape, pre-machining and post-machining imaging steps, and comparing light intensity differences between corresponding pixels in images to identify the kerf, with optional steps for determining optimal imaging regions and detecting positional deviations under varying light conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image processing is used to recognize the kerf on the dicing tape, then the kerf position can be detected, but the adhesive pattern on the dicing tape interferes with accurate recognition
Solution Approach 1:
The patent extracts the harmful adhesive pattern information from the image processing by using subtraction processing. The pre-machining image (containing the pattern) is subtracted from the post-machining image, removing the pattern interference and leaving only the kerf information for accurate recognition.
Solution Approach 2:
The patent performs preliminary imaging of the dicing tape pattern before machining to create a reference image. This pre-machining image is then used in subtraction processing to eliminate the pattern interference, enabling accurate kerf recognition without the harmful factors.
2Ease of operation
If the kerf is formed in the surplus region of the workpiece, then alignment can be performed, but the kerf may be formed in the device region causing positioning errors
Solution Approach 1:
The patent uses image processing to detect the actual kerf position and provides feedback for alignment correction. By comparing the detected kerf position with the reference line, the system calculates positional deviation and adjusts the cutting blade position accordingly, enabling precise alignment regardless of initial kerf location.
3Ease of operation
If coarse alignment is performed by cutting into the dicing tape, then the reference line and kerf can be aligned, but the adhesive pattern interferes with accurate kerf recognition
Solution Approach 1:
The patent replaces mechanical alignment methods with optical image processing. Instead of relying solely on mechanical positioning, the system uses imaging to detect the kerf position and calculates alignment corrections, substituting mechanical operations with optical measurement and computational analysis for higher precision.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate recognition of the kerf, reducing errors in alignment and facilitating precise cutting operations by distinguishing the kerf from the tape pattern, thus improving the accuracy of hairline alignment.
Implementation Method 1
a pre-machining imaging step of imaging an optimal region of the dicing tape where the workpiece is not bonded
Implementation Method 2
a post-machining imaging step of imaging the optimal region with the kerf formed therein
Data Source
AI summary
A recognition method of a kerf includes a bonding step of bonding a workpiece to a dicing tape greater in size than the workpiece, a pre-machining imaging step of imaging an optimal region of the dicing tape where the workpiece is not bonded, a kerf forming step of forming a kerf in the optimal region by a cutting machine, a post-machining imaging step of imaging the optimal region with the kerf formed therein, and a recognition step of comparing intensities of light received at each two corresponding pixels in respective images of the optimal region as acquired by the pre-machining imaging step and the post-machining imaging step, subtracting the each two pixels where intensities of received light are the same, and recognizing as the kerf a region formed by the remaining pixels.