Abstract: Systems and methods of imaging a sample using a tilted charged-particle beam. The apparatus may comprise a first deflector located between the charged-particle source and an objective lens and configured to deflect the charged-particle beam away from the primary optical axis; a second deflector located substantially at a focal plane of the objective lens and configured to deflect the charged-particle beam back towards the primary optical axis; and a third deflector located substantially at a principal plane of the objective lens, wherein the third deflector is configured to shift a wobbling center of the objective lens to an off-axis wobbling location, and wherein the first and the second deflectors are configured to deflect the charged-particle beam to pass through the off-axis wobbling location to land on a surface of a sample at a first landing location and having a beam-tilt angle.

Abstract: An apparatus includes a first charged particle beam manipulator positioned in a first layer configured to influence a charged particle beam and a second charged particle beam manipulator positioned in a second layer configured to influence the charged particle beam. The first and second charged particle beam manipulators may each include a plurality of electrodes having a first set of opposing electrodes and a second set of opposing electrodes. A first driver system electrically connected to the first set may be configured to provide a plurality of discrete output states to the first set. A second driver system electrically connected to the second set may be configured to provide a plurality of discrete output states to the second set. The first and second charged-particle beam manipulators may each comprise a plurality of segments; and a controller having circuitry configured to individually control operation of each of the plurality of segments.

Abstract: Systems and methods of providing a probe spot in multiple modes of operation of a charged-particle beam apparatus are disclosed. The method may comprise activating a charged-particle source to generate a primary charged-particle beam and selecting between a first mode and a second mode of operation of the charged-particle beam apparatus. In the flooding mode, the condenser lens may focus at least a first portion of the primary charged-particle beam passing through an aperture of the aperture plate to form a second portion of the primary charged-particle beam, and substantially all of the second portion is used to flood a surface of a sample. In the inspection mode, the condenser lens may focus a first portion of the primary charged-particle beam such that the aperture of the aperture plate blocks off peripheral charged-particles to form the second portion of the primary charged-particle beam used to inspect the sample surface.

Abstract: Systems and methods of providing a probe spot in multiple modes of operation of a charged-particle beam apparatus are disclosed. The method may comprise activating a charged-particle source to generate a primary charged-particle beam and selecting between a first mode and a second mode of operation of the charged-particle beam apparatus. In the flooding mode, the condenser lens may focus at least a first portion of the primary charged-particle beam passing through an aperture of the aperture plate to form a second portion of the primary charged-particle beam, and substantially all of the second portion is used to flood a surface of a sample. In the inspection mode, the condenser lens may focus a first portion of the primary charged-particle beam such that the aperture of the aperture plate blocks off peripheral charged-particles to form the second portion of the primary charged-particle beam used to inspect the sample surface.

Abstract: The present disclosure discloses a machine learning-based method for defending a voice assistant from being controlled by an inaudible command, including following steps: 1) collecting data of positive and negative samples, 2) performing data segmentation on data of the positive and negative samples; 3) selecting and normalizing sample features; 4) selecting a classifier to be trained and generate a detection model for a malicious voice command; 5) detecting a voice command to be detected by the detection model. The present disclosure selects an original feature selection method, and for smart devices of different types, it is necessary to obtain normal voice commands and malicious voice commands by means of a smart device of this type, and use them as the positive and negative samples to train a specific classifier for the device. Such a customized approach can well solve a problem that detection and defense between devices cannot work.

Abstract: The disclosure discloses a method and a system for generating image adversarial examples based on an acoustic wave. The method includes: acquiring an image containing a target object or a target scene; generating simulated image examples for the acquired image, wherein the simulated image examples have adversarial effects on a deep learning algorithm in a target machine vision system; optimizing the generated simulated image examples to obtain an optimal adversarial example and corresponding adversarial parameters; and injecting the adversarial parameters into an inertial sensor of the target machine vision system in a manner of an acoustic wave, such that the adversarial parameters are used as sensor readings that will cause an image stabilization module in the target machine vision system to operate to generate particular blurry patterns in a generated real-world image so as to generate an image adversarial example in a physical world.

Abstract: Systems and methods of imaging a sample using a charged-particle beam apparatus are disclosed. The charged-particle beam apparatus may include a compound objective lens comprising a magnetic lens and an electrostatic lens, the magnetic lens comprising a cavity, and an electron detector located immediately upstream from a polepiece of the magnetic lens and inside the cavity of the magnetic lens. In some embodiments, deflectors may be located between the electron detector and the opening of the polepiece adjacent to the sample to achieve a large field of view. Electron distributions among the detectors can be manipulated without changing the landing energy by changing the potential of the control electrode(s) in the electrostatic objective lens. The electron source can be operated with several discrete potentials to cover different landing energies, while the potential difference between electron source and the extractor is fixed.

Abstract: Systems and methods of providing a probe spot in multiple modes of operation of a charged-particle beam apparatus are disclosed. The method may comprise activating a charged-particle source to generate a primary charged-particle beam and selecting between a first mode and a second mode of operation of the charged-particle beam apparatus. In the flooding mode, the condenser lens may focus at least a first portion of the primary charged-particle beam passing through an aperture of the aperture plate to form a second portion of the primary charged-particle beam, and substantially all of the second portion is used to flood a surface of a sample. In the inspection mode, the condenser lens may focus a first portion of the primary charged-particle beam such that the aperture of the aperture plate blocks off peripheral charged-particles to form the second portion of the primary charged-particle beam used to inspect the sample surface.

Abstract: The present disclosure discloses a machine learning-based method for defending a voice assistant from being controlled by an inaudible command, including following steps: 1) collecting data of positive and negative samples, 2) performing data segmentation on data of the positive and negative samples; 3) selecting and normalizing sample features; 4) selecting a classifier to be trained and generate a detection model for a malicious voice command; 5) detecting a voice command to be detected by the detection model. The present disclosure selects an original feature selection method, and for smart devices of different types, it is necessary to obtain normal voice commands and malicious voice commands by means of a smart device of this type, and use them as the positive and negative samples to train a specific classifier for the device. Such a customized approach can well solve a problem that detection and defense between devices cannot work.

Abstract: Provided is anti-interference method and circuit, and a mobile terminal. The method includes: eliciting a target interference signal to be suppressed from an interference source; performing phase reversal processing on the elicited target interference signal; performing amplitude modulation processing according to a target amplitude value of the target interference signal to obtain an interference suppression signal, where the interference suppression signal and the target interference signal to be suppressed are opposite in phase and the interference suppression signal has an amplitude value less than or equal to the target amplitude value; and suppressing the target interference signal in the interference source using the obtained interference suppression signal.

Abstract: Provided is anti-interference method and circuit, and a mobile terminal. The method includes: eliciting a target interference signal to be suppressed from an interference source; performing phase reversal processing on the elicited target interference signal; performing amplitude modulation processing according to a target amplitude value of the target interference signal to obtain an interference suppression signal, where the interference suppression signal and the target interference signal to be suppressed are opposite in phase and the interference suppression signal has an amplitude value less than or equal to the target amplitude value; and suppressing the target interference signal in the interference source using the obtained interference suppression signal.