3300/20-12-01-01-00-01Wu Yongjian, vice president of Tencent Cloud and head of Tencent Cloud intelligent production research, said that Tencent Cloud has been deeply engaged in the field of industrial quality inspection for many years, and has built an AI visual inspection product matrix including industrial quality inspection training platform TI-AOI and Tencent Cloud TI platform. For a long time, Tencent Cloud has focused on providing advanced software and algorithms, and is open to the majority of partners for integration, and jointly created solutions covering the entire process of industrial quality inspection, helping industry customers to improve the efficiency and accuracy of product defect identification.
For example, Rongqi Technology and Tencent Cloud for the wireless charging module project, the two sides jointly polish the technology, the detection accuracy rate quickly increased to more than 98%, can meet the customer’s mass production index of more than 99%, the two sides also actively promote cooperation in lithium battery, consumer electronics and other fields, to help more vertical areas to solve quality inspection problems. Visu Technology is also working with Tencent Cloud to explore business expansion and innovation in the field of automotive glass defect detection.
After years of accumulation, Tencent Cloud and partners have accumulated a wealth of AI quality inspection solutions in more than 20 industries such as 3C, lithium, photovoltaic, semiconductor, and automotive, and a single enterprise has completed more than 20 million product appearance tests, creating a series of benchm3300/20-12-01-01-00-01arking projects such as Fuchi High-tech, and the industrial AI quality inspection ecology is thriving.
The upgraded industrial quality inspection training platform TI-AOI is a zero-code development and delivery tool for industrial visual quality inspection scenarios. It takes deep learning inspection as the core to build an efficient and stable data processing and workflow. TI-AOI version 2.3 introduces innovative features such as deep learning precision segmentation, unsupervised anomaly detection, and further improves inference performance. At the same time, the new version has also achieved remarkable results in cutting-edge technologies such as defect sensing universal model and small sample visual prompting, and its excellent performance has been verified by multiple quality inspection scenario POC.
To do a good job in industrial AI quality inspection projects, it is necessary to have the system engineering capability of hardware and software integration of “optical, mechanical, electrical, software and computing”. Over the years, Tencent Cloud and its partners in the field of industrial AI quality inspection have made products, found scenes, made benchmarks, and pushed copies, achieved phased results, and established a healthy cooperation model of deep cooperation and complementary advantages.
The establishment of the industrial AI Quality inspection Ecological Alliance is the further deepening of Tencent Cloud industrial AI quality inspection ecology. By building an ecological platform, it will further promote partner exchanges and cooperation, widely absorb quality inspection partners, explore the formulation of industry-level no3300/20-12-01-01-00-01rms and standards, build an ecosystem of resource sharing, interconnection and mutual win, and sharing results, and then promote continuous innovation and iteration of industrial AI quality inspection.