Cheng Uei Precision Industry Co., Ltd.

Wistron Corporation

Tsai-sheng Shen

Chih-Wei Cheng

Kuang-Yu Wang

Han-Wei Wang

Han-wei Wang

Jung-kuang Liu

Jung-Kuang Liu

An image-recognition method is provided. The method includes the following steps: receiving structured data, wherein the structured data includes training-set data and testing-set data, and the structured data includes a plurality of groups, and each group includes one or more types, and each type includes a plurality of check-point images; training an artificial-intelligence (AI) model using the training-set data; inputting the testing-set data into the AI model to obtain a model evaluation of the AI model; and determining one or more first types with a lower overall recognition rate or a lower confidence level in the structured data, and deletes or corrects the check-point images in the one or more first types to update the structured data.

Image-recognition apparatus, image-recognition method, and non-transitory computer-readable storage medium thereof

An image-recognition method is provided. The method includes the following steps: receiving a plurality of check-point images, and classifying the check-point images into a plurality of groups; classifying the check-point images in each group into a plurality of types to generate first structured data; balancing a number of the check-point images in each type of each group in the first structured data to generate second structured data, wherein the second structured data includes training-set data and testing-set data; training an artificial-intelligence (AI) model using the training-set data; and inputting the testing-set data into the AI model to obtain a model evaluation of the AI model.

An image-recognition method is provided. The method includes the following steps: receiving a plurality of check-point images, and classifying the check-point images into a plurality of groups; classifying the check-point images in each group into a plurality of types to generate first structured data, wherein the first structured data includes a first layer and a second layer, and the first layer indicates the groups in different statuses, the second layer is located with directories of the first layer, and the types in each group of the second layer indicate different components in a status corresponding to each group; and balancing a number of the check-point images in each type of each group in the first structured data to generate second structured data, wherein the second structured data is used to train an AI model for image recognition.

A surface treatment equipment is designed for forming nickel barriers on a plurality of terminals for preventing solder wicking is disclosed. The surface treatment equipment includes a retractable feeding system, a laser engraving system, an image sensor, and a control system. The retractable feeding system is utilized to transmit a strip that has the terminals. The laser engraving system is utilized to ablate the terminals. The image sensor is utilized to collect a plurality of images of the ablated terminals. The control system receives the images to perform image recognition. When a defective terminal is recognized, the control system controls the retractable feeding system to transmit in reverse and controls the laser engraving system to repeatedly ablate the defective terminal. The defective terminals can be automatically recognized by the image sensor accompanying the control system. Thus, the drawback of a human visual inspection is solved.

Surface treatment equipment including a laser engraving system for treatment of a strip

Provided is a method for measuring the heights of components based on laser ranging, including the following steps of: 1) a laser ranger uniformly moving along a test route and measuring the distance from each test point to the laser ranger; 2) the laser ranger transmitting the measurement result and the measurement time to a storage unit; 3) the storage unit sending the received measurement result and the received measurement time to an analysis unit; and 4) according to the data sent by the storage unit and referring to the test route and the position of each test point, the analysis unit calculating the height of each test point and outputting it. The method of the present invention omits the time of determining the position of each point and the start-stop time of the laser ranger at each test point so that saving the large amount of the measurement time.

Method for measuring the heights of components based on laser ranging

A method for adjusting and testing an image sensor module is provided. The method of the present invention includes steps of: (1) calculating longitudinal and transverse deviation values of a center point of a lens relative to a standard reference position specified on a lens barrel, and an angle deviation value of the lens according to position signals of the lens on the image sensor module; (2) compensating the longitudinal and transverse deviation values of the center point of the lens, and the angle deviation value of the lens for exactly positioning the lens; and (3) adjusting focus of the image sensor module. The method of the present invention can compensate the mechanism errors produced in the mounting process of the image sensor module for conveniently and exactly adjusting the focus of the image sensor module. An apparatus for adjusting and testing an image sensor module is also provided.

Method and apparatus for adjusting and testing an image sensor module

An automatic gluing mechanism includes a vertical board and a support rack and a power cylinder mounted to the vertical board. The mechanism includes a position adjusting member, a cam, two glue applicator tubes, a first connection block, two slidable members, a resilient element, and a stop block. The position adjusting member is slidably coupled to the vertical board and has an end coupled to the power cylinder and another end slidably coupled to the first connection block. The two slidable members are slidably mounted to the first connection block. The two glue applicator tubes are rotatably mounted to the two slidable members and connected to each other by the resilient element. The cam is mounted to the position adjusting member and forms a camming surface. The slidable members engage and follow the camming surface. The stop block is mounted to the support rack to stop the first connection block.

Growing community of inventors

Taipei, Taiwan

Tsai-sheng Shen