Abstract: An optical inspection method for an optical inspection device comprising an optical lens is provided according to an embodiment of the disclosure. The optical inspection method includes: obtaining a first image of an object by the optical lens; performing an edge detection on the first image to obtain a second image comprising an edge pattern; and performing a defect inspection operation on the second image based on a neural network architecture to inspect a defect pattern in the second image. In addition, an optical inspection device and an optical inspection system are provided according to embodiments of the disclosure.

Abstract: A defect inspection system, connected to an automatic visual inspection device, is provided, including the followings. A re-inspection server (VRS) receives a defect image and a defect location. A training terminal stores trained modules. A classification terminal receives the defect image and the defect location, reads a target trained module corresponding to the defect image, classifies the defect image according to the target trained module to obtain a labeled defect image, and sends the labeled defect image to the VRS. A re-inspection terminal receives the labeled defect image from the VRS, and sends a verified operation corresponding to the labeled defect image to the VRS. A labeling re-inspection terminal receives the verified operation and the labeled defect image, and a labeling result corresponding to the labeled defect image. The VRS sends the labeling result and the labeled defect image to the training terminal to train a corresponding training module.

Abstract: The present invention is an automated optical inspection method using deep learning, comprising the steps of: providing a plurality of paired image combinations, wherein each said paired image combination includes at least one defect-free image and at least one defect-containing image corresponding to the defect-free image; providing a convolutional neural network to start a training mode of the convolutional neural network; inputting the plurality of paired image combinations into the convolutional neural network, and adjusting a weight of at least one fully connected layer of the convolutional neural network through backpropagation to complete the training mode of the convolutional neural network; and performing an optical inspection process using the trained convolutional neural network.

Abstract: The present disclosure provides an image-based classification system, comprising an image capturing device, and a processing device connected to the image capturing device. The image capturing device is used for capturing an image of an object. The object has a surface layer and an inner layer. The processing device is configured to use a deep learning model, perform image segmentation on the image of the object, define a surface-layer region and an inner-layer region of the image, and generate classification information. The surface-layer region and the inner-layer region correspond respectively to the surface layer and the inner layer of the object.

Abstract: A semiconductor defects inspection apparatus for inspection of bubble defects of an object is provided. The semiconductor defects inspection apparatus includes a carrier, an optical system, an infrared image capturing device, and a processing unit. The carrier is adapted for bearing the object. The optical system provides an illumination beam to the object to produce an image beam. The infrared image capturing device is disposed on a transmission path of the image beam. The infrared image capturing device is adapted for receiving the image beam to be transformed into an image information. The processing unit is electrically connected to the infrared image capturing device and adapted for analyzing the object according to the image information.

Abstract: A semiconductor defects inspection apparatus for inspection of bubble defects of an object is provided. The semiconductor defects inspection apparatus includes a carrier, an optical system, an infrared image capturing device, and a processing unit. The carrier is adapted for bearing the object. The optical system provides an illumination beam to the object to produce an image beam. The infrared image capturing device is disposed on a transmission path of the image beam. The infrared image capturing device is adapted for receiving the image beam to be transformed into an image information. The processing unit is electrically connected to the infrared image capturing device and adapted for analyzing the object according to the image information.

Abstract: An optical inspection method for an optical inspection device comprising an optical lens is provided according to an embodiment of the disclosure. The optical inspection method includes: obtaining a first image of an object by the optical lens; performing an edge detection on the first image to obtain a second image comprising an edge pattern; and performing a defect inspection operation on the second image based on a neural network architecture to inspect a defect pattern in the second image. In addition, an optical inspection device and an optical inspection system are provided according to embodiments of the disclosure.

Abstract: A defect inspection system, connected to an automatic visual inspection device, is provided, including the followings. A re-inspection server (VRS) receives a defect image and a defect location. A training terminal stores trained modules. A classification terminal receives the defect image and the defect location, reads a target trained module corresponding to the defect image, classifies the defect image according to the target trained module to obtain a labeled defect image, and sends the labeled defect image to the VRS. A re-inspection terminal receives the labeled defect image from the VRS, and sends a verified operation corresponding to the labeled defect image to the VRS. A labeling re-inspection terminal receives the verified operation and the labeled defect image, and a labeling result corresponding to the labeled defect image. The VRS sends the labeling result and the labeled defect image to the training terminal to train a corresponding training module.

Abstract: The present invention is an automated optical inspection method using deep learning, comprising the steps of: providing a plurality of paired image combinations, wherein each said paired image combination includes at least one defect-free image and at least one defect-containing image corresponding to the defect-free image; providing a convolutional neural network to start a training mode of the convolutional neural network; inputting the plurality of paired image combinations into the convolutional neural network, and adjusting a weight of at least one fully connected layer of the convolutional neural network through backpropagation to complete the training mode of the convolutional neural network; and performing an optical inspection process using the trained convolutional neural network.

Abstract: The present invention relates to an eye searching method and an eye state detection apparatus and an eye searching apparatus using the same. The method comprises the steps of: defining each center point of two nostrils from a face image; computing a distance D between the two center points of the two nostrils and a midpoint between the two center points of the two nostrils; determining a reference point that has a horizontal distance k1×D and a vertical distance k2×D spaced apart from the midpoint, wherein k1=1.6˜1.8, k2=1.6˜1.8; and defining a rectangle around the reference point as its center, wherein the rectangle has a width and a length greater than the width, and the rectangle surrounds the eye of the face image. Accordingly, the present invention can quickly and accurately find out a certain part of the eye from the face image.

Abstract: A device for generating signal by detecting facial movement and operating method thereof is provided, which includes: an image capture unit, and a processing unit. The image capture unit for obtaining an image series. The processing unit receives the images series from the image capture unit, wherein the processing unit includes an image background removal module, an image extracting module, and a comparator, wherein the image background removing module processes each of the image series respectively to obtain a facial image, wherein the feature location module determines a location of a pair of nostrils in the facial image, defines a mouth searching frame, and acquires a data of mouth movements through the mouth searching frame, wherein the comparator compares the data of mouth movements with predetermined facial information, and generates a designated signal according to the comparison result.

Abstract: A device for generating signal by detecting facial movement and operating method thereof is provided, which includes: an image capture unit, and a processing unit. The image capture unit for obtaining an image series. The processing unit receives the images series from the image capture unit, wherein the processing unit includes an image background removal module, an image extracting module, and a comparator, wherein the image background removing module processes each of the image series respectively to obtain a facial image, wherein the feature location module determines a location of a pair of nostrils in the facial image, defines a mouth searching frame, and acquires a data of mouth movements through the mouth searching frame, wherein the comparator compares the data of mouth movements with predetermined facial information, and generates a designated signal according to the comparison result.

Abstract: A monitor is utilized for installing inside an automobile to monitor a number of passengers in an automobile. The monitor includes an image capturing device, and a processing unit is connected to the image capturing device. The image capturing device successively captures a plurality of images in the automobile. The processing unit receives the images from the image capturing device to arrange an image sequence, wherein the processing unit includes an image dividing module, an image processing module, and a motion detecting module. The image dividing module divides each of the images into a plurality of recognition blocks, the image processing module compares the recognition blocks of consecutive images in the image sequence to acquire an image variance value, and the motion detecting module has a threshold value and determines the recognition blocks as a human-present condition when the image variance value surpasses the threshold value.

Abstract: A signal triggering method and an in-vehicle electronic apparatus are provided. A plurality of images of a driver is continuously captured by using an image capturing unit, and a face motion information or an eyes open-shut information is obtained by detecting a face motion or an eyes open/shut action of the driver through the images. When the face motion information or the eyes open-shut information matches a threshold information, a specific signal is triggered and transmitted to a specific device.

Abstract: The present invention relates to an eye searching method and an eye state detection apparatus and an eye searching apparatus using the same. The method comprises the steps of: defining each center point of two nostrils from a face image; computing a distance D between the two center points of the two nostrils and a midpoint between the two center points of the two nostrils; determining a reference point that has a horizontal distance k1×D and a vertical distance k2×D spaced apart from the midpoint, wherein k1=1.6˜1.8, k2=1.6˜1.8; and defining a rectangle around the reference point as its center, wherein the rectangle has a width and a length greater than the width, and the rectangle surrounds the eye of the face image. Accordingly, the present invention can quickly and accurately find out a certain part of the eye from the face image.

Abstract: An eye-controlled communication system comprises an eye controlled aid and a visiting aid. The eye controlled aid has an eye controlled module, a first display module and a first processing unit. The eye controlled module detects eye movements of patient, generates a control command based on the detected results and then transmits to the first processing unit. The first processing unit bases the control command to allow patient to operate the first operator interface by using eye movements. The first processing unit generate an execute command according to the operating results, and performs the execute command, wherein the execute command includes information of transmitting messages to the second processing unit. The second processing unit receives the execute command and performs the execute command. Thus, the system of the present invention can help patient to easily express himself or herself to his or her family members or friends.

Abstract: A two-way communication system by using eye movements of the present invention comprises a first display module, a second display module and a data processing module. The first display module is configured to display first operative window interface and to detect the eye movements of the users to provide various operation and control. The second display module is configured to display user information, which allows the users to communicate with others immediately and conveniently. The data processing module is electrically connected with the first display module and the second display module, which responses each operation command and transmits the information. The first and second display modules can be adjusted freely for users' convenient review and operation.

Abstract: An ear hanging type eye control device includes a main body, an ear hook, and an image-taking module. The main body is provided therein with the electronic components necessary for operation. The ear hook extends outward from the main body so as to hook over one of the user's ears. The image-taking module is connected to the main body and is configured for taking images of the user's eyes. The ear hanging type eye control device is capable of detecting and identifying eyeball movements and executing a variety of control instructions.

Abstract: The present invention provides a device and a method for detecting a driver's attention through interactive voice questioning, which are used to monitor a head action of the driver to determine an attention level of the driver, and mainly operate in a process of asking a question by the device, answering the question by the driver, and making a determination by the device. The device includes an image acquiring unit configured to acquire an image of the driver, a processing unit configured to perform image digital processing on the image, a voice questioning unit configured to give a voice questioning instruction, and a safety determining unit configured to make a value comparison and a determination according to a feature of the image. An alert message will be given to warn the driver if the image fails to comply with requirements of the voice questioning instruction.

Abstract: An in-car eye control method is provided to monitor a car driver's eye movements, thereby allowing the driver to communicate with people outside the car or control in-car equipment. The in-car eye control device essentially includes an image capturing unit for capturing an image of the driver, an processing unit for performing operations on the image taken, an eye determination unit for monitoring the driver's eye movements and generating an eye control instruction, and a system operation unit for executing a system process according to the eye control instruction. Thus, the in-car eye control method provides a diversity of communication or control solutions.