Abstract: The described aspects and implementations enable efficient calibration of a sensing system of a vehicle. In one implementation, disclosed is a method and a system to perform the method, the system including the sensing system configured to collect sensing data, characterizing an environment of the vehicle, the sensing data including infrared sensing data. The system further includes a data processing system operatively coupled to the sensing system and configured to process the sensing data using a classifier machine-learning model to obtain a classification of one or more vulnerable road users present in the environment of the vehicle.

Abstract: A device and method for traffic monitoring and a non-transitory tangible machine-readable medium for use in the device are disclosed. The device stores a probability model. The device records a packet quantity transmitted by a device under test in a monitoring time period. The device determines that the device under test is in an abnormal state when a probability of occurrence corresponded to by the packet quantity and the monitoring time period is lower than a probability threshold, wherein the probability of occurrence is determined by the probability model.

Abstract: A device and method for traffic monitoring and a non-transitory tangible machine-readable medium for use in the device are disclosed. The device stores a probability model. The device records a packet quantity transmitted by a device under test in a monitoring time period. The device determines that the device under test is in an abnormal state when a probability of occurrence corresponded to by the packet quantity and the monitoring time period is lower than a probability threshold, wherein the probability of occurrence is determined by the probability model.

Abstract: A computer-implemented method for generating a unified machine learning model using a neural network on a data processing apparatus is described. The method includes the data processing apparatus determining respective learning targets for each of a plurality of object verticals. The data processing apparatus determines the respective learning targets based on two or more embedding outputs of the neural network. The method also includes the data processing apparatus training the neural network to identify data associated with each of the plurality of object verticals. The data processing apparatus trains the neural network using the respective learning targets and based on a first loss function. The data processing apparatus uses the neural network trained to generate a unified machine learning model, where the model is configured to identify particular data items associated with each of the plurality of object verticals.

Abstract: A packet analysis apparatus, method, and non-transitory computer readable medium thereof are provided. The packet analysis apparatus stores a plurality of packets whose formats are unknown. The packet analysis apparatus calculates a plurality of cross-correlation values of the packets. The packet analysis apparatus decides at least one group according to the cross-correlation values and at least one first threshold, wherein each group includes a subset of the packets. The packets included in a specific group of the groups define a plurality of bit positions. Each packet included in the specific group has a plurality of bits. For each of the bit positions, the packet analysis apparatus calculates a variation degree of the bits corresponding to the bit positions. The packet analysis apparatus selects the at least one bit position whose variation degree(s) is/are smaller than a second threshold as at least one field boundary of the specific group.

Abstract: The present disclosure illustrates an image sensor. The image sensor includes an image sensing array and a voltage supply array. The image sensing array and the voltage supply array are coupled to an analog-to-digital converter array. The image sensing array captures image data. The image sensing array supports one of a rolling shutter mechanism and a global shutter mechanism according to setting. The voltage supply array includes a plurality of voltage supply circuits to supply dummy voltage. During an auto-zero period, the voltage supply array provides the dummy voltage to the analog-to-digital converter array. Pluralities of comparators of the analog-to-digital converter array execute an auto-zero function based on the dummy voltage. After finishing the auto-zero function, the image sensing array outputs the image data to the analog-to-digital converter array. The analog-to-digital converter array makes the image data be digital.

Abstract: A packet analysis apparatus, method, and non-transitory computer readable medium thereof are provided. The packet analysis apparatus stores a plurality of packets whose formats are unknown. The packet analysis apparatus calculates a plurality of cross-correlation values of the packets. The packet analysis apparatus decides at least one group according to the cross-correlation values and at least one first threshold, wherein each group includes a subset of the packets. The packets included in a specific group of the groups define a plurality of bit positions. Each packet included in the specific group has a plurality of bits. For each of the bit positions, the packet analysis apparatus calculates a variation degree of the bits corresponding to the bit positions. The packet analysis apparatus selects the at least one bit position whose variation degree(s) is/are smaller than a second threshold as at least one field boundary of the specific group.

Abstract: A method for reducing fixed pattern noise of an image sensor is provided. The method includes: accessing pixel data of at least one test frame in a test environment; and calculating/deriving average values of each column based on at least one portion of pixel data of the column in the at least one test frame, wherein the average values of columns are used as calibration values for calibrating image pixel data of columns when the image sensor operates in a normal light source environment.

Abstract: Systems and methods are disclosed to provide an Advanced Warning System (AWS) for a driver of a vehicle, by capturing traffic scene types from a single camera video; generating real-time monocular SFM and 2D object detection from the single camera video; detecting a ground plane from the real-time monocular SFM and the 2D object detection; performing dense 3D estimation from the real-time monocular SFM and the 2D object detection; generating a joint 3D object localization from the ground plane and dense 3D estimation; and communicating a situation that requires caution to the driver.

Abstract: The present invention provides an image sensor circuit with a power noise filtering function and a control method thereof. The image sensor circuit includes: an image sensing unit, for sensing an image, to generate an image sensed signal; a power noise sensing unit, for sensing a power noise, to generate a power noise signal; a differential amplifier circuit, which is coupled to the image sensing unit and the power noise sensing unit, for receiving the image sensed signal and the power noise signal, to generate a differential filtered signal by a differential amplification operation between the image sensed signal and the power noise signal; and an analog-to-digital converter (ADC) circuit, which is coupled to the differential amplifier circuit, for receiving the differential filtered signal, to generate a digital image signal.

Abstract: The present disclosure illustrates an image sensor. The image sensor includes an image sensing array and a voltage supply array. The image sensing array and the voltage supply array are coupled to an analog-to-digital converter array. The image sensing array captures image data. The image sensing array supports one of a rolling shutter mechanism and a global shutter mechanism according to setting. The voltage supply array includes a plurality of voltage supply circuits to supply dummy voltage. During an auto-zero period, the voltage supply array provides the dummy voltage to the analog-to-digital converter array. Pluralities of comparators of the analog-to-digital converter array execute an auto-zero function based on the dummy voltage. After finishing the auto-zero function, the image sensing array outputs the image data to the analog-to-digital converter array. The analog-to-digital converter array makes the image data be digital.

Abstract: A method for reducing fixed pattern noise of an image sensor is provided. The method includes: accessing pixel data of at least one test frame in a test environment; and calculating/deriving average values of each column based on at least one portion of pixel data of the column in the at least one test frame, wherein the average values of columns are used as calibration values for calibrating image pixel data of columns when the image sensor operates in a normal light source environment.

Abstract: A method for inducing thrombosis in an artery of a living animal includes: implanting an implantable medical device into a living animal, the implantable medical device including a light-emitting element and a wireless power receiver; administrating a photo-sensitizing dye material into the living animal; and providing wirelessly a control signal, from an external power controller, to the wireless power receiver which transmits electrical power to the light-emitting element in response to the control signal to drive the light-emitting element to emit light for irradiating the artery of the living animal.

Abstract: The present invention provides an image sensor circuit with a power noise filtering function and a control method thereof. The image sensor circuit includes: an image sensing unit, for sensing an image, to generate an image sensed signal; a power noise sensing unit, for sensing a power noise, to generate a power noise signal; a differential amplifier circuit, which is coupled to the image sensing unit and the power noise sensing unit, for receiving the image sensed signal and the power noise signal, to generate a differential filtered signal by a differential amplification operation between the image sensed signal and the power noise signal; and an analog-to-digital converter (ADC) circuit, which is coupled to the differential amplifier circuit, for receiving the differential filtered signal, to generate a digital image signal.

Abstract: The disclosure is a device and a method for Non Return to Zero (NRZ) Clock Data Recovery (CDR) calibration, which includes a CDR unit and a weight calculator unit. The CDR unit receives a compensative signal of an equalization filter to generate an error signal, a sampling clock signal, a transition sampling signal and a data signal. The weight calculator unit receives the error signal, the transition sampling signal and the data signal, and then uses a run length technique to generate weight data. The weight data controls a voltage control oscillator (VCO) which calibrates the phase and the frequency of the sampling clock signal.

Abstract: Systems and methods are disclosed to provide an Advanced Warning System (AWS) for a driver of a vehicle, by capturing traffic scene types from a single camera video; generating real-time monocular SFM and 2D object detection from the single camera video; detecting a ground plane from the real-time monocular SFM and the 2D object detection; performing dense 3D estimation from the real-time monocular SFM and the 2D object detection; generating a joint 3D object localization from the ground plane and dense 3D estimation; and communicating a situation that requires caution to the driver.

Abstract: A structure of semiconductor device includes a first semiconductor layer; an intermediate layer on a surface of said first semiconductor layer; a second semiconductor layer on said intermediate layer, wherein said intermediate layer and said second semiconductor layer are integrated to a set of sub-structures; and a semiconductor light emitting device on said second semiconductor layer.

Abstract: A multilayer printed circuit board structure is formed by stacking an aluminum foil substrate, a first pre-impregnated body, an aluminum foil middle layer, a second pre-impregnated body and a copper foil surface layer sequentially. Both the first pre-impregnated body and the second pre-impregnated body are composed by a fiber cloth impregnated with a heat conduction material in order that the heat conduction material can fill up the gaps of the fiber cloth. The heat conduction material is mixed from at least a resin and a filling material.

Abstract: A lens assembly includes a first lens including a first convex surface facing the object side, a first concave surface with an aperture stop facing the image side; a second lens including a second convex surface facing the object side and a second concave surface facing the image side; a first lens further including. Defining that f, f1, f2 are a focal length of the lens assembly, the first lens, and the second lens; D is a distance between the first lens and the second lens; TTL is a total track length; R1 is a radius of the first convex surface; R2 is a radius of the first concave surface; R3 and R4 are respectively radius of the second convex surface and the second concave surface; and ? is the image circle; the following conditions are satisfied: 0.01<|f/f2|<0.2; 0.1<R1/f1<0.5; 0.8<f1/TTL<1.1; 0.2<D/f<0.5; 0.3<R2/?<0.7; and 0.03<(R3?R4)/(R3+R4)<0.3.

Abstract: Network interface apparatus with power management is disclosed, which comprises a physical layer circuit, for receiving a packet on a network; a media access control circuit, for performing the media access processing on the packet to output a processed packet; an interface circuit, coupled to the media access control layer, for transmitting the processed packet to a bus; a detecting circuit, coupled to the physical layer circuit, for detecting a transmitting status of the packet on the network to output a detecting signal; a loading control circuit, coupled to the detecting circuit, for controlling a load positioned in the interface circuit according to the detecting signal.