Abstract: A liveness detection system uses a combination of interactive and silent tests to verify physical presence of a live person at a client device. The system superimposes a box on a live video of the space in front of the client device and instructs the user to move to a position that places his or her face within the box. While the user's face is within the box, the system captures a sequence of frames from the video while an amount of illumination cast on the user's face is switched between bright and dark levels according to a random, semi-random, or predefined flash sequence. The system analyzes the resulting sequence of frames to confirm that the user's head pose is consistent with the location of the box, and that changes in brightness of the image of the user's face across the captured frames are consistent with the flash sequence.

Abstract: Methods and systems are presented for detecting physical tampering on a document based on analyzing an image of the document. When the image of the document is obtained, multiple contours are identified in the image based on pixel characteristics of the image. Dimension attributes of the contours are determined. Contours that are determined to correspond to borders or texts of the documents based on the dimension attributes are eliminated. A second text detection process based on a polygon method is performed on at least one remaining contour to determine whether the at least one remaining contour links multiple text elements together. The document is determined to have been physically manipulated when at least on contour remains in the image.

Abstract: Aspects of the present disclosure involve systems, methods, devices, and the like for user identification using Artificial Intelligence, Machine Learning, and data analytics. In one embodiment, a verification system and method is introduced that can provide user authentication using parallel modeling for face identification. The verification system used includes a face identification module for use in the identification and verification using parallel processing of a received image with a claimed identity. The parallel processing includes an ensemble of machine learning models processed in parallel for optimal performance.

Abstract: Techniques are disclosed relating to methods that include receiving, by a computer system, a plurality of images of an object taken from a video during which there is relative movement between the object and a camera that captures the video. The method may further include in response to determining that the video does not include a single image that meets a clarity threshold for the object, creating, by the computer system, a merged image of the object by combining portions of different images of the plurality of images such that the clarity threshold for the object is satisfied by the merged image. The method may also include capturing, by the computer system, information about the object using the merged image.

Abstract: A liveness detection system uses a combination of interactive and silent tests to verify physical presence of a live person at a client device. The system superimposes a box on a live video of the space in front of the client device and instructs the user to move to a position that places his or her face within the box. While the user's face is within the box, the system captures a sequence of frames from the video while an amount of illumination cast on the user's face is switched between bright and dark levels according to a random, semi-random, or predefined flash sequence. The system analyzes the resulting sequence of frames to confirm that the user's head pose is consistent with the location of the box, and that changes in brightness of the image of the user's face across the captured frames are consistent with the flash sequence.

Abstract: Systems, methods, and computer program products for determining an attack on a neural network. A data sample is received at a first classifier neural network and at a watermark classifier neural network, wherein the first classifier neural network is trained using a first dataset and a watermark dataset. The first classifier neural network determines a classification label for the data sample. A watermark classifier neural network determines a watermark classification label for the data sample. A data sample is determined as an adversarial data sample based on the classification label for the data sample and the watermark classification label for the data sample.

Abstract: Systems and/or techniques for facilitating image forgery detection via pixel-metadata consistency analysis are provided. In various embodiments, a system can receive an electronic image from a client device. In various cases, the system can obtain a pixel vector and/or an image metadata vector that correspond to the electronic image. In various aspects, the system can determine whether the electronic image is authentic or forged, based on analyzing the pixel vector and the image metadata vector via at least one machine learning model.

Abstract: Systems, methods, and computer program products for determining an attack on a neural network. A data sample is received at a first classifier neural network and at a watermark classifier neural network, wherein the first classifier neural network is trained using a first dataset and a watermark dataset. The first classifier neural network determines a classification label for the data sample. A watermark classifier neural network determines a watermark classification label for the data sample. A data sample is determined as an adversarial data sample based on the classification label for the data sample and the watermark classification label for the data sample.

Abstract: Systems and/or techniques for facilitating image forgery detection via headpose estimation are provided. In various embodiments, a system can receive a document from a client device. In various cases, the system can identify, by executing a first trained machine learning model, an object that is depicted in the document. In various instances, the system can determine, by executing a second trained machine learning model, a pose of the object. In various aspects, the system can determine, by executing a third trained machine learning model, whether the document is authentic or forged based on the pose of the object. In various embodiments, the system can, in response to determining that the document is forged, transmit an unsuccessful validation message to the client device.

Abstract: The present invention discloses a method for simulating stochastic oscillation in an individual-granularity long-distance expressway traffic volume using a quantum harmonic oscillator, which includes: firstly, describing the speed and position of a vehicle by a quantum superposition state, and constructing an energy eigenequation of the quantum harmonic oscillator to represent movement of the vehicle; secondly, constructing an n-order Hermite polynomial based on the energy eigenequation, constructing a quantum harmonic oscillator model for simulating the stochastic oscillation in the long-distance traffic flow in a mode featuring aliasing of multiple strategies, and optimizing a solution model; and finally, constructing a mapping mechanism between the probability and the traffic volume to simulate the traffic volume.

Abstract: Techniques are disclosed relating to determining whether document objects included in an image correspond to known document types. In some embodiments, a computing system maintains information specifying a set of known document types. In some embodiments, the computing system receives an image that includes objects. In some embodiments, the computing system analyzes, using a first neural network, the image to identify a document object and location information specifying a location of the document object within the image. In some embodiments, the computing system determines, using a second neural network, whether the document object within the image corresponds to a document type specified in the set of known document types, where the determining is performed based on the location information of the document object. In some embodiments, disclosed techniques may assist in automatically extracting information from documents, which in turn may advantageously decrease processing time for onboarding new customers.

Abstract: Provided is a method for controlling an air conditioner provided with a bipolar ion generation module. The method comprises the following steps: an air conditioner controller determining whether a purification mode control signal for controlling a bipolar ion generation module to start operation is received; if the purification mode control signal is received, the air conditioner controller detecting whether an air conditioner is in a working state; if the air conditioner is in the working state, controlling the bipolar ion generation module to start working; and if the air conditioner is in a standby state, outputting a first drive signal to drive a fan to start operation, and at the same time, controlling a first timer to start timing, and when the timing of the first timer satisfies a first set duration, controlling the bipolar ion generation module to start working. Further disclosed is an air conditioner.

Abstract: Systems, methods, and computer program products for determining an attack on a neural network. A data sample is received at a first classifier neural network and at a watermark classifier neural network, wherein the first classifier neural network is trained using a first dataset and a watermark dataset. The first classifier neural network determines a classification label for the data sample. A watermark classifier neural network determines a watermark classification label for the data sample. A data sample is determined as an adversarial data sample based on the classification label for the data sample and the watermark classification label for the data sample.

Abstract: Techniques are disclosed relating to determining whether document objects included in an image correspond to known document types. In some embodiments, a computing system maintains information specifying a set of known document types. In some embodiments, the computing system receives an image that includes objects. In some embodiments, the computing system analyzes, using a first neural network, the image to identify a document object and location information specifying a location of the document object within the image. In some embodiments, the computing system determines, using a second neural network, whether the document object within the image corresponds to a document type specified in the set of known document types, where the determining is performed based on the location information of the document object. In some embodiments, disclosed techniques may assist in automatically extracting information from documents, which in turn may advantageously decrease processing time for onboarding new customers.

Abstract: Aspects of the present disclosure involve systems, methods, devices, and the like for user identification using Artificial Intelligence, Machine Learning, and data analytics. In one embodiment, a verification system and method is introduced that can provide user authentication using parallel modeling for face identification. The verification system used includes a face identification module for use in the identification and verification using parallel processing of a received image with a claimed identity. The parallel processing includes an ensemble of machine learning models processed in parallel for optimal performance.

Abstract: The disclosure provides for compositions and methods of controlling and/or regulating plant size in a controlled environment, for example, a greenhouse or environment for the purpose of testing plants. In another aspect, the disclosure provides for methods of dwarfing plant growth by utilizing a composition or method described herein. The disclosure further provides for methods described herein wherein a plant growth regulator is used to stunt plant growth in order enable more consistent and/or accurate plant feature and property testing.

Abstract: An air duct machine including a cabinet, an electric control box, a heat exchanger and a centrifugal fan; the electric control box, the heat exchanger and the centrifugal fan are all disposed inside the cabinet, one of side panels of the cabinet is provided with a return air vent and an air outlet, the electric control box is disposed close to the return air vent, the centrifugal fan is disposed close to the air outlet, and the heat exchanger is disposed between the electric control box and the centrifugal fan. This improves the selectivity and flexibility of installation of the air duct machine, and facilitates maintenance. The air duct machine can have an ultra-thin design when installed in the ceiling or wall, so that the installation space is reduced. Moreover, the airflow direction will not change due to the change of the installation method.

Abstract: The invention provides genes that activate wax biosynthesis in plants. Also provided are constructs comprising these sequences, plants transformed therewith and methods of use thereof. The invention allows the modification of plants for increased wax production, particularly in leaves. The inventors have demonstrated increased drought tolerance in connection with increased leaf wax production engineered in plants.

Abstract: The invention provides genes that activate wax biosynthesis in plants. Also provided are constructs comprising these sequences, plants transformed therewith and methods of use thereof. The invention allows the modification of plants for increased wax production, particularly in leaves. The inventors have demonstrated increased drought tolerance in connection with increased leaf wax production engineered in plants.