Microdefect integration early warning method and device

Abstract

The invention relates to a microdefect integration early warning method and a device which uses the method. A microscopic inspection machine is used in sequence to detect a substrate to be detected, defects on the substrate to be detected are found and positioned and are stored into a memory unit, a data collection unit collects various pieces of detection defect data of the substrate to be detected from the memory unit, and the detection defect data is transmitted into an integration classification unit. The integration classification unit simultaneously imports a defect database, historicaldefects are analyzed and compared with an existing defect so as to judge a defect situation on the substrate to be tested, and classification integration is carried out. When necessary, when the defect achieves a certain extent of injury, an early warning mechanism is triggered to work. By use of the method, the deficiency that detection personnel manually carry out manual analysis on the detection defect data each time is omitted, and the operation state of a generation field is reflected in real time. By use of the device which uses the method, detection efficiency is improved, and cost waste brought by detection result lagging is avoided.

Description

technical field [0001] The invention relates to the field of liquid crystal panel manufacturing, in particular to the quality monitoring of the manufacturing process of liquid crystal screens. Background technique [0002] The liquid crystal industry uses a microscopic inspection machine as a substrate and product microscopic defect inspection and data storage has a high technical level, and the missed inspection rate is very low. On the one hand, it can detect the surface process or substrate after coating / exposure / development of the substrate Micro-defects and defects on the back, so as to make a quality judgment on the substrate, so as to repair or scrap the defects of the substrate before the next process, so as to avoid microscopic defects of the product (light leakage, foreign objects in the box, scratches, etc.) and waste of components ; Another function is to store substrate defect data and pictures. These data can not only help engineers realize automatic management...

AI Technical Summary

Problems solved by technology

[0003] However, the currently used microscopic inspection machine completes the presentation of defect data through optical scanning and photographing technology, thereby outputting the data to the storage unit, uploading the production line production data to the data collection system through the network, and then downloading the data by personal terminals; Or through on-site manual detection and collection of real-time data, but the integration and classification of data still requires engineering processing before the identification of defects and the monitoring and timely response to abnormal points in production, but often the data in production does not pass System to achieve integration, classification and real-time engineering early warning, so there is time difference and cumbersome workload for production monitoring and management

[0004] Therefore, although we already have relatively complete optical scanning technology and imaging technology, the data processing and statistics are relatively backward, and engineering needs to spend a lot of energy and time to interpret the data and artificially analyze the data. Integration and classification, for a highly competitive industry, efficient production efficiency and abnormal handling efficiency are the most effective cost control, but the current equipment does not fully utilize the data statistics, integration, and classification systems to bring Advantages in engineering and production