Abstract: An elbow joint rehabilitation system and an elbow joint rehabilitation method are provided. The elbow joint rehabilitation system includes a support member, a motor, a torque sensing unit, a first electromyography sensor, a second electromyography sensor and a motor control device. In the elbow joint rehabilitation method, the support member is configured to support an arm of a patient. Thereafter, the torque sensing unit is configured to sense the torque applied on the support member to obtain a sensed arm torque signal. Then, the first electromyography sensor and the second electromyography sensor are configured to sense the muscle activities of biceps and triceps of the patient to obtain electromyography signals. Thereafter, the motor is controlled to drive the support member to perform rehabilitation in accordance with the sensed arm torque signal, the electromyography signals and a current support member position provided by the motor.

Abstract: An adaptive controller used in a photoplethysmography sensing system, comprises a plurality of hardware circuits which are configured to: receive a photoplethysmography signal (hereinafter, “PPG signal”) processed; determine whether the PPG signal processed satisfies with a requirement; output the PPG signal processed if the PPG signal processed satisfies with a requirement; and adjust a gain of an amplifier for amplifying the PPG signal and/or a driving signal of a light source if the PPG signal processed does not satisfy with a requirement.

Abstract: A solar module assembly includes a frame having an upper portion encompassing an area and a mid portion disposed below the upper portion. A plurality of solar panels is arranged in a string, sandwiched between two transparent panes forming a single string panel. The solar panels occupy less than the area of the upper portion. Each of the plurality of solar panels has a pair of opposing edges. A reflector is mounted on the mid portion to reflect light selectively.

Abstract: An image analyzing method is provided and includes: extracting a first feature vector according to global information of a digital image; dividing the digital image into multiple regions, and inputting each region into a convolutional neural network to obtain a second feature vector; merging the first feature vector with the second feature vectors to obtain a third feature vector; and performing an image analyzing process according to the third feature vector.

Abstract: A set of 3D user-designed images is used to create a high volume of realistic scenes or images which can be used for training and testing deep learning machines. The system creates a high volume of scenes having a wide variety of environmental, weather-related factors as well as scenes that take into account camera noise, distortion, angle of view, and the like. A generative modeling process is used to vary objects contained in an image so that more images, each one distinct, can be used to train the deep learning model without the inefficiencies of creating videos of actual, real life scenes. Object label data is known by virtue of a designer selecting an object from an image database and placing it in the scene. This and other methods care used to artificially create new scenes that do not have to be recorded in real-life conditions and that do not require costly and time-consuming, manual labelling or tagging of objects.

Abstract: A system for process control of a composite fabrication process comprises an automated composite placement head, a vision system, and a computer system. The automated composite placement head is configured to lay down composite material. The vision system is connected to the automated composite placement head and configured to produce image data during an inspection of the composite material, wherein the inspection takes place at least one of during or after laying down the composite material. The computer system is configured to identify inconsistencies in the composite material visible within the image data, and make a number of metrology decisions based on the inconsistencies.

Abstract: An optical mechanism includes a substrate, a supporting pad, an optical element and a fixed member. The substrate has an opening and a recess, the recess surrounds the opening and has a bottom surface. The supporting pad is disposed in the recess and has a convex curved portion, the convex curved portion protrudes away from the bottom surface of the recess, and a gap exists between the convex curved portion and the bottom surface. The optical element is disposed on the substrate, and the optical element covers the opening and abuts against the convex curved portion. The fixed member is fixed on the substrate and is adapted to press the optical element onto the substrate. The optical mechanism and the projection device of the invention can effectively prevent the optical element from being broken.

Abstract: A bipolar junction transistor includes: an emitter region; a base region; and a collector region, wherein each of the emitter region, the base region, and the collector region comprises fin-shaped structures. Preferably, the emitter region, the base region, and the collector region are disposed along a first direction and the fin-shaped structures are disposed along a second direction, in which the first direction is orthogonal to the second direction.

Abstract: Disclosed is a class of new aryl-phosphorus-oxygen compounds as shown in formula (I) as EGFR kinase inhibitors, and pharmaceutically acceptable salts thereof.

Abstract: An image analyzing method is provided and includes: extracting a first feature vector according to global information of a digital image; dividing the digital image into multiple regions, and inputting each region into a convolutional neural network to obtain a second feature vector; merging the first feature vector with the second feature vectors to obtain a third feature vector; and performing an image analyzing process according to the third feature vector.

Abstract: A method for measuring spasticity is provided and includes: obtaining sensing signals corresponding to a limb movement through at least one sensor during a period of time; transforming the sensing signals into a two-dimensional image; and inputting the two-dimensional image into a convolutional neural network to output a spasticity determination result.

Abstract: An apparatus to facilitate instruction scheduling is disclosed. The apparatus includes one or more processors to receive a block of instructions, divide the block of instructions into a plurality of sub-blocks based on a register pressure bounded by a predetermined threshold and instructions in each of the plurality of sub-blocks for processing.

Abstract: A highly active mixed transition metal oxide material has been developed. The material may be sulfided to generate metal sulfides which are used as a catalyst in a conversion process such as hydroprocessing. The hydroprocessing may include hydrodenitrification, hydrodesulfurization, hydrodemetallation, hydrodesilication, hydrodearomatization, hydroisomerization, hydrotreating, hydrofining, and hydrocracking.

Abstract: A highly active quaternary mixed transition metal oxide material has been developed. The material may be sulfided to generate metal sulfides which are used as a catalyst in a conversion process such as hydroprocessing. The hydroprocessing may include hydrodenitrification, hydrodesulfurization, hydrodemetallation, hydrodesilication, hydrodearomatization, hydroisomerization, hydrotreating, hydrofining, and hydrocracking.

Abstract: An image recognition method is provided and includes: decoding a digital image file to obtain a digital image; providing, by a front end application, a user interface so that a user interacts with the user interface through a browser, in which the user interface renders the digital image; receiving an image editing operation corresponding to the digital image through the user interface; obtaining characteristic information corresponding to a sample region of the digital image from the digital image file according to the image editing operation; transmitting, by the front end application, the characteristic information to a server to perform an image recognition procedure.

Abstract: A medical image modeling system and a medical image modeling method are provided. The medical image modeling system includes an input device, a processing device, and a storage device. The input device obtains medical image data. The processing device is coupled to the input device. The processing device converts the medical image data into a bone model in response to a first operational command. The storage device is coupled to the processing device. The storage device stores a model database. The processing device reads the model database in response to a second operational command to obtain a bone reconstruction block corresponding to a defect block of the bone model. The processing device integrates a bone reconstruction model depicted corresponding to the bone reconstruction block into the defect block of the bone model to generate an integrated bone model.

Abstract: A gate-controlled bipolar junction transistor includes a substrate, an emitter region, a base region disposed on one side of the emitter region, and a collector region disposed on one side of the base region and being opposite to the emitter region. The emitter region includes first fin structures, first metal gates extending across the first fin structures, and an emitter contact plug on the first fin structures. A gate contact region is disposed between the emitter region and the base region. Each of the first metal gates includes an extended contact end portion protruding toward the base region. The extended contact end portion is disposed within the gate contact region. A gate contact is disposed on the extended contact end portion.

Abstract: An aircraft includes a propulsion system, a sensor system, a control system, and a processing system including a memory and a processor. The memory is configured to store computer-executable instructions. The processor is configured to access the memory and to execute the computer-executable instructions to perform the following steps: obtaining a set of first weights of a processing unit of a neural network; ternarizing each weight included in the set of first weights to obtain a set of second weights; generating an output of the processing unit based on the set of second weights and a set of inputs of the processing unit; and training weights included in the set of first weights of the processing unit of the neural network based on an error cost function including an error term and a structurally sparse term.

Abstract: A set of 3D user-designed images is used to create a high volume of realistic scenes or images which can be used for training and testing deep learning machines. The system creates a high volume of scenes having a wide variety of environmental, weather-related factors as well as scenes that take into account camera noise, distortion, angle of view, and the like. A generative modeling process is used to vary objects contained in an image so that more images, each one distinct, can be used to train the deep learning model without the inefficiencies of creating videos of actual, real life scenes. Object label data is known by virtue of a designer selecting an object from an image database and placing it in the scene. This and other methods care used to artificially create new scenes that do not have to be recorded in real-life conditions and that do not require costly and time-consuming, manual labelling or tagging of objects.

Abstract: A method for fabricating bipolar junction transistor (BJT) includes the steps of: providing a substrate having an emitter region, a base region, and a collector region; performing a first implantation process to form a first well region in the base region; and performing a second implantation process to form a second well region in the emitter region. Preferably, the first well region and the second well region comprise different concentration.