:::

[2022 Solution] AI electric eyes replace human eyes. Huiyan Intelligence uses AI to help the manufacturing industry with quality control

In response to a small number of diverse customer needs in the manufacturing industry, there is an urgent need to find AI solutions from the cloud to the terminal. Huiyan Intelligence provides software and hardware integration solutions - BailAI image inspection solutions to assist traditional manufacturing industries in improving process efficiency and product quality, and achieving the initial goal of transformation.

After the government declared 2017 to be Taiwan's "Year of AI", new AI startups have sprung up in Taiwan. Huiyan Intelligence, established in 2018, targets smart manufacturing and provides a platform for AI image analysis and process optimization. Deep learning methods detect product defects and assembly step abnormalities, assisting companies in building infrastructure from terminals to the cloud, so that factory production can be automatically monitored to improve the efficiency and quality of the manufacturing process.

Familiar with the production line quality control process and using AI image detection as the main focus of entrepreneurship

Liu Yawen, founder and CEO of Huiyan Intelligence, was young. She entered the manufacturing industry after graduating from university and held a quality control position in the plastic injection process of hard disk parts. "At that time, she was already on the production line, and for the production machine "She is very familiar with the production line process." She later switched careers to marketing planning, and then worked as an AI product manager. When the time came, Liu Yawen decided to start a business, focusing on AI image recognition in the manufacturing industry.

"The difficulty for enterprises is the lack of AI development teams. Even if there is an AI team, development projects will take a lot of time, at least 6-12 months." Liu Yawen, who is well versed in the market's pain points, said that the problem the platform needs to solve is to provide traditional manufacturing Employees in the industry who do not need a program development background can also build their own AI model platform to remotely assist in troubleshooting production lines and subsequent system maintenance operations, helping companies save development time and labor costs.

BailAI image detection platform usage scenarios.

▲BailAI image detection platform usage scenarios

Faced with many competitors providing AI image recognition in the market, what are the technical advantages of Huiyan Intelligence? Liu Yawen said that at this stage, many companies have AOI optical inspection equipment, but the bottleneck in the application of AOI optical inspection is that it can only be used for defect inspection with fast production lines and large quantities, and each inspection or production must be readjusted. parameter. According to her understanding of the industry, limited by the fact that AOI equipment often costs over one million Taiwan dollars, most small and medium-sized traditional manufacturing industries do not have strong financial resources, but they also want to do automated testing. This is Huiyan Intelligence. Opportunity.

Liu Yawen went on to say that it is impossible for traditional manufacturing industries to maintain a technical team including AI engineers, data engineers, cloud architects, terminal architecture engineers and other professional talents. Huiyan Intelligence is good at software and hardware integration. Enterprises can use BailAI The image inspection platform can easily solve inspection problems on the production line. In other words, customers only need to provide images or samples to Huiyan’s intelligent training model, deployment model and system integration, and they can easily use AI technology to optimize and monitor production line processes.

Participated in the AI+ Rising Talents Selection Competition and achieved a recognition rate of over 90% in assembly behavior image recognition

For example, a certain connector manufacturer has only 1-2 AI engineers in its technical team. The main problem to be solved is that most operators are on the production line, while quality control and senior managers are at the remote end. The company wants to understand the actual situation of the production line through remote monitoring. Huiyan Intelligent uses industrial cameras to capture production line images and transmits AI image analysis to the remote end. Supervisors and quality control personnel can observe through the screen whether there are any errors in the production line assembly, such as whether the connector heads and lines are connected properly. And other issues.

Huiyan Intelligent's AI image inspection rack operates on Microsoft's Azure cloud platform, and will also be placed around the inspection station through terminal equipment, such as NVIDIA's edge computing equipment. Through a cloud-to-end integrated solution, Assist traditional manufacturing industries to improve production line efficiency and detect problems early. At this stage, Huiyan Intelligence’s customer base includes aviation, electronic peripherals, connectors, metal and other related industries.

2. The accuracy rate reaches more than 90%.

▲ Assembly process solution for human behavior recognition in assembly lines, with an accuracy rate of over 90%

In order to demonstrate the depth of technology, Huiyan Intelligent participated in the 2021 AI+ Rising Stars Selection Event of the Industrial Bureau of the Ministry of Economic Affairs and provided Lite-On Technology with a "Human Behavior Recognition Assembly Process for Assembly Production Line" solution, through camera and AI image recognition. To identify the effective working hours and ineffective working hours of production line operators, that is, to judge the assembly behavior of operators through image recognition of hand posture and position, the accuracy can reach more than 90%.

Liu Yawen added that since the assembly process of electronic components is complex and mostly relies on human labor and cannot be replaced by robotic arms, Huiyan Intelligent used the camera to film the assembly process of the operators at Lite-On's assembly station. The algorithm is then trained and corrected based on the video. The final trained model can directly determine whether there are any errors in the assembly process to improve the overall process.

Introducing the BailAI image detection platform, the project development time is expected to be shortened to 1 month

Since its establishment more than three years ago, Huiyan Intelligence has accumulated a lot of project experience and hopes to commercialize the project experience. Liu Yawen pointed out that the BailAI image detection trial version will be completed this year (2022). Customers can customize the detection of objects based on the precision of the object. Select industrial cameras, video cameras, or even X-rays to capture images, and then automatically label the images through the platform. Huiyan Intelligence will provide AI application models that fit the field for customers to use, and can also make inferences on cloud terminals. , which is convenient for the manufacturing industry to go online. Including metal casings, connectors, electronic peripherals, and mechanical parts in the metal industry and industrial computers, the platform can be used for defect detection and object identification.

At this stage, Huiyan Intelligence will continue to improve its technical capabilities, accumulate customer experience to complete productization, and at the same time accelerate the application of AI detection. In the mid-term, it will build terminal cloud infrastructure and reduce the development time of enterprise AI projects from 6 to 12 months. It is shortened to 1 month, lowering the enterprise usage time and usage threshold. The long-term goal is to target the Southeast Asian market where Taiwanese businessmen gather, expand software and hardware integrated AI solutions to overseas markets, and expand the scale of operations.

Hui Yan Intelligent Participation Winner of the 2021 AI+ Talent Selection Competition of the Industrial Bureau of the Ministry of Economic Affairs

▲Huiyan Intelligent participated in the 2021 AI+ Rising Talent Selection Event of the Industrial Bureau of the Ministry of Economic Affairs and won the award

「Translated content is generated by ChatGPT and is for reference only. Translation date:2024-05-19」

Recommend Cases

這是一張圖片。 This is a picture.
【2023#15】South Estate Precision Marketing & APP 2.0 Integration Project

Industry Accommodation and Food Services IMajor Category Accommodation Industry 55 Subcategory Industry Pain Points Significant seasonal demand fluctuations Some areas face reduced visitors and declining performance during the off-season Tourism experience and quality management Providing a high-quality tourism experience is a crucial goal for the industry, yet sometimes faces challenges with unstable service quality and customer complaints High demand for human resources Human resources are critical to the industry, but sometimes there is a shortage of workforce and insufficient skills, necessitating enhanced training and talent development Digital transformation pressure With the advancement of digitalization and technology, the industry needs to address challenges associated with digital transformation, online sales, marketing strategies, and technological innovation Benefits of AI Adoption Personalized services AI can analyze and understand the preferences, needs, and behavior patterns of visitors, thereby providing personalized travel recommendations and services, improving visitor satisfaction, increasing repeat visitation and loyalty, and facilitating business growth Smart interaction AI can act as an intelligent customer service assistant, such as virtual assistants or chatbots, answering visitor questions, providing immediate support, and customizing interactions based on visitor needs, enhancing the customer experience Accurate demand forecasting AI can analyze visitor data and market trends to forecast demand peaks and changes, helping to optimize inventory management, workforce allocation, and pricing strategy, thereby enhancing operational efficiency and revenue Customer feedback insights AI can analyze the reviews and emotional feedback of visitors to extract key information, quickly identify and respond to customer needs, improve service quality, and enhance customer satisfaction and reputation Enriched travel experiences AI can integrate technologies such as augmented reality and virtual reality to provide richer and more immersive travel experiences, attract visitors, add value to the travel offerings, and gain competitive advantages CommonAITechnologies or Applications Random Forest, K-means Clustering, ContentCollaborative Recommendations AI applications such as the Random Forest algorithm analyze large amounts of visitor data to learn and predict their preferences and behavior for demand forecasting AI like K-means clustering and contentcollaborative recommendation algorithms segregate customers, offering more accurate and personalized travel experiences and services Generative AI Using natural language processing techniques to analyze and understand visitor communications including website reviews, social media posts, coupled with AI like generative pre-trained transformers for automated dialogue responses, sentiment analysis, and opinion mining better understand and respond to visitor needs Convolutional Neural Networks Utilizing AI such as convolutional neural network algorithms to identify and analyze content in images and videos, such as visitor facial expressions and actions, to understand visitor reactions and experience, and to further improve service quality and travel arrangements 「Translated content is generated by ChatGPT and is for reference only Translation date:2024-05-19」

這是一張圖片。 This is a picture.
Shin Kong Mitsukoshi Introduces Yuhui AI Search Engine in 2023

Industry Wholesale and Retail Business GMajor Category Retail Subcategory 47 Industry Pain Points High market competition pressure Businesses must face challenges from traditional physical stores and online e-commerce to gain competitive advantages, requiring continuous optimization in pricing, product diversity, shopping experience, and service quality Slow pace of digital transformation With the rapid progress of technology, consumer shopping patterns have undergone significant changes, such as the rise of e-commerce reshaping the traditional retail industry Department stores should carefully integrate online and offline sales channels to ensure a consistent and superior shopping experience for consumers Inventory management and supply chain need integration Businesses need to manage large inventories precisely to ensure smooth supply and rapid product delivery Lack of personalized shopping experience Generation Z, digital natives, particularly value the shopping experience, including the usability and search functions of platforms and apps Insufficient information security protection With the increase in massive data and online transactions, protecting consumers' personal data and payment security has become a primary task for businesses IntroductionAIBenefits Personalized consumer recommendations By analyzing consumers' purchase history, preferences, and behavior patterns, businesses can provide personalized product recommendations, enhancing consumer satisfaction and purchase intent Virtual fitting and realistic experiences Using technologies like virtual fitting rooms and virtual reality to provide consumers with realistic shopping experiences online This can reduce the time spent in physical fitting rooms and decrease the demand for them Enhancing cross-selling success rates Based on consumers' purchase records and behavior patterns, offering relevant cross-selling suggestions can increase the shopping frequency and enhance transaction value Virtual assistants and automated response services By deploying virtual assistants, chatbots, and automated response systems, businesses can provide consumers with real-time customer support, while also effectively saving on human resource costs Consumer market insights and trends By analyzing large volumes of data, businesses can gain consumer insights and market trends, helping to optimize merchandise display, marketing strategies, and inventory management, and further enhancing sales performance and profits Common AI technologies or applications Random Forest, Recurrent Neural Networks With AI algorithms such as random forests or recurrent neural networks, by analyzing historical data such as browsing, clicking, and purchasing, businesses can identify important consumer trends and patterns, and capture consumer behavior patterns over time, helping to more accurately predict future consumer trends Collaborative filtering, content filtering Using AI techniques like content filtering or collaborative filtering, consumer behavior and preferences can be analyzed to provide personalized product recommendations Recurrent Neural Networks, Natural Language Processing, Generative AI Employing AI technologies such as recurrent neural networks combined with natural language processing for voice recognition and natural language understanding, deriving consumer insights and emotional tendencies also using generative AI like generative pre-trained transformers to develop virtual assistants or chatbots, providing real-time customer support and personalized services, available 247 「Translated content is generated by ChatGPT and is for reference only Translation date:2024-05-19」

這是一張圖片。 This is a picture.
Briefing sessions on application for the 2024 AI implementation subsidies in Northern, Central, and Southern Taiwan will start on March 26

Enterprise AI power, the real power for digital development of the AI industry Sign up now for the Administration for Digital industries 2024 AI Subsidy Briefing Session Taiwan is facing the challenges and opportunities brought by the digital economy To accelerate the upgrading and transformation of industries through AI, let us give you a hand This project accepts applications for industry AI implementation and diffusion, in order to accelerate the implementation of AI applications and technologies in industries It encourages companies to produce common solutions based on industry AI technology needs and implementation readiness, combined with appropriate data collection and analysis It helps match innovative AI technologies with industry needs, and gathers together capabilities of AI startups, information service providers, and various industries With hardware pushing the economy and software pulling it forward, the capabilities of international partners are utilized to gradually make industries more intelligent and implement the AI industry, so as to realize the vision of a smart nation Application for industry AI implementation and diffusion has begun To get started with the subsidy application process, you cannot miss this briefing session The maximum subsidy for each company is NT4 million SubjectsAI startupsinformation service providerstechnology companiesfields that required AI Event information The Central and Southern Taiwan events are offline events, only the Taipei event will be held online simultaneously Date Session Method Time Meeting location March 26, 2024 Tuesday Southern Taiwan Offline 13301430 Registration at 1300 Conference Room on the 1st floor of the Southern Taiwan Industry Promotion Center1F, No 6, Fuxing 4th Road, Qianzhen District, Kaohsiung City March 27, 2024 Wednesday Central Taiwan Offline 14001500Registration at 1330 Tai Chung Computer Association-903 meeting room9th Floor, No 95-8, Sec 1, Changping Rd, Beitun District, Taichung City March 29, 2024 Friday Northern Taiwan Offline online video conference 14001500Registration at 1330 Conference Room C, 6F, PCBC6F, No 99, Fuxing N Rd, Songshan District, Taipei City nbsp Event Agenda Date Time Duration Agenda Speaker March 26, 2024 Tuesday 13001330 30 minutes Registration Time nbsp 13301335 5 minutes Opening nbsp 13351345 10 minutes Project description and sharing of guidance resources Service team representative 13451405 20 minutes Industry AI Implementation Verification Plan and Diffusion Application Instructions Taipei Computer Association 14051430 25 minutes General discussion and exchanges QampA All March 27, 2024 Wednesday 13301400 30 minutes Registration Time nbsp 14001405 5 minutes Opening nbsp 14051415 10 minutes Project description and sharing of guidance resources Service team representative 14151435 20 minutes Industry AI Implementation Verification Plan and Diffusion Application Instructions Taipei Computer Association 14351500 25 minutes General discussion and exchanges QampA All March 29, 2024 Friday 13301400 30 minutes Registration Time nbsp 14001405 5 minutes Opening nbsp 14051415 10 minutes Project description and sharing of guidance resources Service team representative 14151435 20 minutes Industry AI Implementation Verification Plan and Diffusion Application Instructions Taipei Computer Association 14351500 25 minutes General discussion and exchanges QampA All nbsp Please click on the following URL to register now and learn more detailed information and online websiteRegistration websitehttpsseminarstcaorgtwD15r00761aspx Event ContactTaipei Computer Association 02-2570-6337Miss Gao 9512 Miss Hung 9827 Miss Lin 9865