Abstract: A method for detecting defects in products from images thereof and an electronic device applying the method inputs a defect image repair data set into an autoencoder to train the autoencoder, and generates a reconstructed image, calculates a reference error value between the sample image and the reconstructed image by a preset error function, and set a threshold value based on the reference error value. The electronic device inputs an image possibly revealing a defect into the autoencoder and generates the reconstructed image corresponding to the image to be detected, and uses the preset error function to calculate the reconstruction error between the image and the reconstructed image, thereby determining whether the image being analyzed does reveal defects. When the reconstruction error is greater than the threshold value, a determination is made that a defect is revealed.

Abstract: A method for recognizing different object-categories within images based on texture classification of the different categories, which is implemented in an electronic device, includes extracting texture features from block images segmented from original images according to at least one Gabor filter; determining a grayscale level co-occurrence matrix of each block image according to the texture features; calculating texture feature statistics of each block image according to the grayscale level co-occurrence matrix; training and generating an object recognition model using the texture features and the texture feature statistics; and recognizing and classifying at least one object in original image according to the object recognition model.

Abstract: An acoustic metasurface structure is configured to absorb sounds. The acoustic metasurface structure comprises a main body, an externally-connecting configuration and an inner configuration. The externally-connecting configuration and the inner configuration are respectively formed inside the main body. An externally-connecting tube of the externally-connecting configuration is in fluid communication with an external environment and an externally-connecting cavity of the externally-connecting configuration. An inner tube of the inner configuration is in fluid communication with the externally-connecting cavity and an inner cavity of the inner configuration. With the externally-connecting configuration and the inner configuration forming a series-type structure in the main body, the acoustic metasurface structure increases an acoustic impedance.

Abstract: The disclosure describes techniques for configuring wireless routers of a wireless mesh network with one or more secondary wireless access points (WAPs) such that client devices connected to the mesh network may remain online in the event that a primary WAP goes offline or otherwise becomes unavailable. For instance, a user of the wireless mesh network may specify a device, such as a mobile electronic device connected to a cellular network, to operate as a secondary WAP in the event that a primary WAP becomes unavailable. Thereafter, upon a wireless router of the mesh network detecting that the primary WAP is unavailable, the wireless router may perform a network scan to identify SSID(s) that are available within the environment. The wireless router may also instruct the other routers of the mesh network to perform a network scan.

Abstract: A method for neural network model encryption and decryption includes a first apparatus stores a neural network model and obtains hardware configuration information of the first apparatus, and obtains an encryption key accordingly; encrypts the neural network model by a predetermined encryption algorithm; a second apparatus obtains the encrypted neural network model from the first apparatus, transmits a decryption request to the first apparatus, obtains the hardware configuration information from the first apparatus, obtains a decryption key based on the hardware configuration information; and decrypts the encrypted neural network model by a predetermined decryption algorithm.

Abstract: A method for detecting and counting pedestrians in real-time for statistical purposes together with facial recognition of such pedestrians acquires video for analysis. Images showing pedestrians are extracted, such pedestrians being identified by a first detection model and pedestrian frames are outputted. A facial identification operation is executed based on the pedestrian frames and facial sub-frames are extracted. Removal of duplications of individual pedestrians in the images in the facial sub-frames is executed by a second detection model. Path of movement of detected pedestrians is tracked and labeled based on a specified algorithm when the faces of individual pedestrians are not recorded in a database. When the path of movement passes through a specified location, the face-imaged pedestrian is considered a target object, a total number of the target objects is counted. An apparatus applying the method is also disclosed.

Abstract: A semiconductor structure includes a substrate, a piezoelectric layer, and a stress structure. The substrate includes a first surface and a second surface, wherein a portion of the substrate proximal to the first surface defines a diaphragm. The piezoelectric layer is disposed over the first surface of the substrate and surrounds the diaphragm, wherein the piezoelectric layer includes a first portion and a second portion arranged along a periphery of the diaphragm from a top view. The stress structure includes a plurality of dielectric layers disposed over the piezoelectric layer and between the substrate and the piezoelectric layer, and a total thickness of a first portion of the stress structure overlapping the first portion of the piezoelectric layer is different from a total thickness of a second portion of the stress structure overlapping the second portion of the piezoelectric layer. A method for manufacturing a semiconductor structure is also provided.

Abstract: An image augmentation method applied to an electronic device is provided. The method includes constructing a variational learner and a discriminator based on a fully convolutional neural network. A target image is obtained by inputting a gas leakage image into the variational learner. A variational autoencoder model is obtained by training the variational learner based on a discrimination result of the discriminator on the target image. A reconstruction accuracy rate is calculated based on a test image, an augmented model is obtained by adjusting the variational autoencoder model based on the gas leakage image, in response that the reconstruction accuracy rate being less than a preset threshold; and an augmented image is obtained by inputting the image to be augmented into the augmented model.

Abstract: A method for detecting defects in product applied in a computer device inputs an image of a product under test to an automatic encoder to obtain a reconstructed image, and the image is segmented into N image blocks and the reconstructed image is segmented into N image blocks. The computer device associates each of the N testing blocks with one reconstructed blocks according to positions of the N testing blocks in the image and positions of the N reconstructed blocks in the reconstructed image. The computer device further calculates mean square errors between each of the N testing blocks and each of the N reconstructed blocks, and associates each mean square error with each of the N testing blocks, whether the product has defects being determined based on the mean square errors corresponding to each of the N testing blocks.

Abstract: A method for manufacturing a semiconductor structure is provided. The method may include several operations. A piezoelectric capacitor is formed over a substrate, wherein the piezoelectric capacitor includes a metal electrode. An intermediate layer is formed on the metal electrode, and is patterned using a first mask layer as a mask. A metal layer is formed on the intermediate layer, wherein the metal layer electrically connects to the metal electrode. The metal layer is patterned using a second mask layer, wherein the intermediate layer is within a coverage area of the metal layer from a top-view perspective after the patterning of the metal layer. A semiconductor structure thereof is also provided.

Abstract: An image augmentation method applied to an electronic device is provided. The method includes constructing a variational learner and a discriminator based on a fully convolutional neural network. A target image is obtained by inputting a gas leakage image into the variational learner. A variational autoencoder model is obtained by training the variational learner based on a discrimination result of the discriminator on the target image. A reconstruction accuracy rate is calculated based on a test image, an augmented model is obtained by adjusting the variational autoencoder model based on the gas leakage image, in response that the reconstruction accuracy rate being less than a preset threshold; and an augmented image is obtained by inputting the image to be augmented into the augmented model.

Abstract: A method for neural network model encryption and decryption includes a first apparatus stores a neural network model and obtains hardware configuration information of the first apparatus, and obtains an encryption key accordingly; encrypts the neural network model by a predetermined encryption algorithm; a second apparatus obtains the encrypted neural network model from the first apparatus, transmits a decryption request to the first apparatus, obtains the hardware configuration information from the first apparatus, obtains a decryption key based on the hardware configuration information; and decrypts the encrypted neural network model by a predetermined decryption algorithm.

Abstract: A method for augmenting defect sample data thereof includes acquiring a positive sample image and defect category information of a surface of a product; inputting the positive sample image and the defect category information to a generative adversarial network (GAN); and generating defect sample data corresponding to the defect category information. An apparatus and a non-transitory computer readable medium for augmenting defect sample data are also disclosed.

Abstract: A method for recognizing different object-categories within images based on texture classification of the different categories, which is implemented in an electronic device, includes extracting texture features from block images segmented from original images according to at least one Gabor filter; determining a grayscale level co-occurrence matrix of each block image according to the texture features; calculating texture feature statistics of each block image according to the grayscale level co-occurrence matrix; training and generating an object recognition model using the texture features and the texture feature statistics; and recognizing and classifying at least one object in original image according to the object recognition model.

Abstract: A method for detecting defects in products revealed by images of the products inputs images of the flaw-free products into an autoencoder for model training to obtain reconstructed images. The images are further processed to obtain target images. A group of testing errors are obtained by comparing the reconstructed images and the target images. An error threshold is selected from the group of the testing errors according to a specified rule. A to-be-analyzed image is inputted for obtaining a candidate be-analyzed reconstructed image, a candidate be-analyzed target image, and a potential be-analyzed error between the candidate be-analyzed reconstructed image and the candidate be-analyzed target image. A result of the to-be-analyzed image confirms defects existing or defects not existing according to the potential be-analyzed error and the error threshold. A defect detection apparatus, an electronic device, and a non-transitory computer-readable storage medium applying the method are also disclosed.

Abstract: A method for detecting defects in product applied in a computer device inputs an image of a product under test to an automatic encoder to obtain a reconstructed image, and the image is segmented into N image blocks and the reconstructed image is segmented into N image blocks. The computer device associates each of the N testing blocks with one reconstructed blocks according to positions of the N testing blocks in the image and positions of the N reconstructed blocks in the reconstructed image. The computer device further calculates mean square errors between each of the N testing blocks and each of the N reconstructed blocks, and associates each mean square error with each of the N testing blocks, whether the product has defects being determined based on the mean square errors corresponding to each of the N testing blocks.

Abstract: A method for detecting defects in products from images thereof and an electronic device applying the method inputs a defect image repair data set into an autoencoder to train the autoencoder, and generates a reconstructed image, calculates a reference error value between the sample image and the reconstructed image by a preset error function, and set a threshold value based on the reference error value. The electronic device inputs an image possibly revealing a defect into the autoencoder and generates the reconstructed image corresponding to the image to be detected, and uses the preset error function to calculate the reconstruction error between the image and the reconstructed image, thereby determining whether the image being analyzed does reveal defects. When the reconstruction error is greater than the threshold value, a determination is made that a defect is revealed.

Abstract: A method for detecting and counting pedestrians in real-time for statistical purposes together with facial recognition of such pedestrians acquires video for analysis. Images showing pedestrians are extracted, such pedestrians being identified by a first detection model and pedestrian frames are outputted. A facial identification operation is executed based on the pedestrian frames and facial sub-frames are extracted. Removal of duplications of individual pedestrians in the images in the facial sub-frames is executed by a second detection model. Path of movement of detected pedestrians is tracked and labeled based on a specified algorithm when the faces of individual pedestrians are not recorded in a database. When the path of movement passes through a specified location, the face-imaged pedestrian is considered a target object, a total number of the target objects is counted. An apparatus applying the method is also disclosed.

Abstract: Disclosures of the present invention describe a halide semiconductor memristor that is suitable for being as an artificial synapse. The halide semiconductor memristor comprises a first electrode layer, an active layer and a second electrode layer, wherein the active layer comprises a first oxide semiconductor film formed on the first electrode layer, a halide semiconductor film formed on the first oxide semiconductor film, and a second oxide semiconductor film formed on the halide semiconductor film Moreover, a variety of experimental data have proved that, this halide semiconductor memristor is indeed suitable for being adopted as a plurality of artificial synapses that are used in manufacture of a neuromorphic device, and exhibits many advantages, including: capable of being driven by a low operation voltage, having a multi-stage adjustable resistance state, and a wide dynamic range of the switching resistance states.

Abstract: A method of determining a parameter of a pattern transfer process and device manufacturing methods are disclosed. In one arrangement, a method includes obtaining a detected representation of radiation redirected by a structure. The structure is a structure formed by applying a pattern processing to a pattern transferred to an earlier formed structure by a pattern transfer process. The pattern processing is such as to remove one or more selected regions in a horizontal plane of the earlier formed structure to form a pattern in the horizontal plane. The pattern is defined by a unit cell that is mirror symmetric with respect to an axis of mirror symmetry. An asymmetry in the detected representation is determined. The determined asymmetry in the detected representation is used to determine a parameter of the pattern transfer process.