Abstract:
A defect inspection system is disclosed, and comprises a linear light source, N number of cameras, a display device, a tag reader, and a modular electronic device, in which the linear light source, the cameras and the modular electronic device are used for conducting a defect inspection of an article. On the other hand, the display device, the tag reader and the modular electronic device are adopted for conducting in production of at least one labeled example. Therefore, the modular electronic device is allowed to apply a machine learning process to an image classifier under using a training dataset containing the labeled examples, thereby producing at least one new defect recognition model or updating the existing defect recognition model.
Abstract:
A flexible and intuitive system is disclosed, and is disposed to be coupled to at least one camera, at least one robotic arm and a host electronic device of an AVI system. During a normal operation of the system, a configuration parameter setting of the AVI system can be completed after at least one time of robotic arm setting operation and at least one time of camera setting operation are conducted. After that, a plurality of article images acquired from a specific article by the camera are upload to a remote electronic device by the system, such that the remote electronic device utilizes the article images to update (re-train) a defect recognition model. Consequently, after the system integrates the defect recognition model into a defect recognition program that is installed in the host electronic device, the AVI system is hence configured for conducting an appearance inspection of the article.
Abstract:
A method to find symmetries along curved paths in input scenes. The method may detect a curve in an input scene and one or more elements on that curve. The method may define and group points for the one or more element on the curve, and define a centroid for each group. The method may then parameterize a transformation in transformation space between each centroid pair in the input scene. The method may then extract transformation paths by clustering points. The method may create phantom objects in case of mirroring along curved paths to help detect the curved paths.
Abstract:
A semiconductor memory employs the redundancy memory technique and the error correction code technique and method of correcting errors. The method of correcting errors reads data bits and a checking bit from a predetermined unit of a first memory array such as a main memory array, and the data bits are checked based on the checking bit to determine if there is any error. If there is an error in the data bits, the checking bit is used to correct the error and the data bits together with the checking bit are written back to the predetermined unit. If there is still error in the data bits after the read-check-write process is repeated a predetermined number of times, the predetermined unit is marked as a faulty unit and the data bits together with the checking bit are written to a second memory array such as a redundancy memory array.
Abstract:
An automatic target image acquisition and calibration system for application in a defect inspection system is disclosed. During the defect inspection system working normally, the automatic target image acquisition and calibration system is configured to find a recognition structure from an article under inspection, and then determines a relative position and a relative 3D coordinate if the article. Therefore, a robotic arm is controlled to carry a camera to precisely face each of a plurality of inspected surfaces of the article, such that a plurality of article images are acquired by the camera. It is worth explaining that, during the defect inspection of the article, there is no need to modulate an image acquiring height and an image acquiring angle of the camera and an illumination of a light source.
Abstract:
A defect inspection system is disclosed, and comprises a linear light source, N number of cameras, a display device, a tag reader, and a modular electronic device, in which the linear light source, the cameras and the modular electronic device are used for conducting a defect inspection of an article. On the other hand, the display device, the tag reader and the modular electronic device are adopted for conducting in production of at least one labeled example. Therefore, the modular electronic device is allowed to apply a machine learning process to an image classifier under using a training dataset containing the labeled examples, thereby producing at least one new defect recognition model or updating the existing defect recognition model.
Abstract:
A motor controlling system is disclosed. The motor controlling system is integrated in an apparatus having roll-to-roll mechanism, and comprises a linear light source, N number of cameras and a modular electronic device. When a roll of continuous web material is discharged by an unwinding unit of the roll-to-roll mechanism so as to be further transferred to an inspection unit, the cameras are controlled by the modular electronic device to capture a plurality of time-consecutive material image frames. Subsequently, the modular electronic device conducts a defection inspection for a segment of the continuous web material by applying an image process to the material image frames. As a result, in case of there being at least one defect found, the modular electronic device stops a plurality of motors of the apparatus from running, thereby making the segment of the continuous web material be positioned on an inspection platform.
Abstract:
An interactive user feedback system for enhancing the inspection accuracy of an automated visual inspection (AVI) system is disclosed. When working normally, the AVI system acquires an article image from a continuous article transferred by a transfer equipment, and then determines whether there is at least one defect feature existing in the article image or not. Subsequently, the interactive user feedback system enables a display of the electronic device show an inspection report region consisting of M×N sub-regions. As such, the display is enabled to show a zoom-in sub-image containing at least one defect after a sub-region is clicked. Therefore, by viewing the zoom-in sub-image, an inspector can determine whether a defect classification data made by the AVI system is correct or not. If not, the inspector is able to revise the defect classification data through the interactive user feedback system.
Abstract:
A surface inspection system for foil article is disclosed. The surface inspection system comprises a box having a top long narrow opening and a bottom long narrow opening, a bridge interface, a first light source, a second light source, a first modular camera device having a first camera, and a second modular camera device having a second camera. In which, the first light source, the second light source, the first modular camera device, and the second modular camera device all accommodated in the box, and are coupled to a control box through the bridge interface. Particularly, this surface inspection system is allowed to be integrated in an automatic production line of a foil article like electro-forming aluminum foil (also called electronic aluminum foil), so as to achieve an in-line inspection of the surface morphology of the electro-forming aluminum foil.
Abstract:
A surface inspection system for foil article is disclosed. The surface inspection system comprises a box having a top long narrow opening and a bottom long narrow opening, a bridge interface, a first light source, a second light source, a first modular camera device having a first camera, and a second modular camera device having a second camera. In which, the first light source, the second light source, the first modular camera device, and the second modular camera device all accommodated in the box, and are coupled to a control box through the bridge interface. Particularly, this surface inspection system is allowed to be integrated in an automatic production line of a foil article like electro-forming aluminum foil (also called electronic aluminum foil), so as to achieve an in-line inspection of the surface morphology of the electro-forming aluminum foil.