Tsinghua University

Beijing Baidu Netcom Science and Technology Co., Ltd.

Shanghai Xiangzhang Electronics Co., Ltd.

Jing Pei

Luping Shi

Guoqi Li

Rong Zhao

Lei Deng

Zhenzhi Wu

Zheyu Yang

Haizheng Xu

Hao Wang

Meng Xu

Ken Chen

Da Lu

Duanyi Xu

Xiantao Chen

Shuang Wu

Jianshe Ma

Xiaohui Lu

Chenyu Su

Jianmin Liu

Longfa Pan

Yi Ni

Jianping Xiong

Dalin Wu

Embodiments of the present disclosure provide a cone-rod dual-modality neuromorphic vision sensor, including: a first preset quantity of voltage-mode active pixel sensor (APS) circuits and a second preset quantity of current-mode APS circuits, where each of the voltage-mode APS circuits includes a first-type photosensitive device, and each of the current-mode APS circuits includes a second-type photosensitive device. The voltage-mode APS can output a target voltage signal representing light intensity information in a target light signal. The obtained target voltage signal represents the light intensity information with a higher precision, and therefore an image with higher quality can be obtained, that is, the image has a higher signal-noise ratio. The voltage-mode APS can output a specified digital signal representing light intensity gradient information in the target light signal, to ensure performance indicators such as an image dynamic range and a shooting speed of the neuromorphic vision sensor, thereby making the neuromorphic vision sensor more stable and robust.

Cone-rod dual-modality neuromorphic vision sensor

Embodiments of the present disclosure provide a dual-modality neuromorphic vision sensor. A first-type current-mode active pixel sensor (APS) circuit can mimic excitatory rod cells, to perceive light intensity gradient information in a target light signal, thereby improving a dynamic arrange of an image sensed by a neuromorphic, vision sensor and its shooting speed. In addition, a first-type control switch is introduced for each of non-target first-type photosensitive devices, to control the obtained light intensity gradient information, and adjust the dynamic arrange of the image sensed by the neuromorphic vision sensor, thereby adjusting the shooting speed, and realizing a reconfigurable effect A voltage-mode APS can mimic cone cells, to output a target voltage signal representing light intensity information in the target light signal, and perceive the light intensity information in the target light signal. In this way, the obtained light intensity information represented by the target voltage signal has a higher precision, thereby ensuring the image quality.

Dual-modality neuromorphic vision sensor

A computer-implemented method for speech recognition is disclosed. The method includes extracting a feature word associated with location information from a speech to be recognized, and calculating a similarity between the feature word and respective ones of a plurality of candidate words in a corpus. The corpus includes a first sub-corpus associated with at least one user, and the plurality of candidate words include, in the first sub-corpus, a first standard candidate word and at least one first erroneous candidate word. The at least one first erroneous candidate word has a preset correspondence with the first standard candidate word. The method further includes in response to the similarity between the feature word and one or more of the at least one first erroneous candidate word satisfying a predetermined condition, outputting the first standard candidate word as a recognition result based on the preset correspondence.

Method, apparatus, electronic device and storage medium for speech recognition

There is provided a neural network information receiving method and system, and a sending method and system. The receiving method comprises: acquiring a reception initiation time for neuron information (S); receiving rostral neuron information output by rostral neurons (S); acquiring delay information of the rostral neuron information according to the reception initiation time, the rostral neuron information and a delay algorithm (S); and determining composite information output by the rostral neurons according to the rostral neuron information and the delay information (S). The receiving method and system and sending method and system take the delay information into consideration in the output information of neurons, such that the neuron information is able to carry more detailed time-domain information, thus augmenting operation modes of the neurons and enhancing the generalization ability of the whole network.

Neural network information receiving method, sending method, system, apparatus and readable storage medium

The disclosure relates to a self-adaptive leakage value neuron information processing method and system. The method includes: receiving front end pulse neuron output information; reading current pulse neuron information, wherein the current pulse neuron information includes self-adaptive membrane potential leakage information; calculating current pulse neuron output information according to the front end pulse neuron output information and the current pulse neuron information; updating the self-adaptive membrane potential leakage information according to the current pulse neuron output information; outputting the current pulse neuron output information. The self-adaptive leakage value neuron information processing system utilizes self-adaptive membrane potential leakage information to participate in calculation of current pulse neuron output information, and utilizes the calculated current pulse neuron output information to update the self-adaptive membrane potential information to participate in calculation of a next time step, and a good balance between sensitivity and stability of a neural network is achieved.

Self-adaptive threshold neuron information processing method, self-adaptive leakage value neuron information processing method, system computer device and readable storage medium

There are provided a neural network weight discretizing method, system and device, and a computer readable storage medium. The method includes acquiring a weight value range and a number of discrete weight states, the weight value range referring to a range of discrete weight values consisting of a maximum weight value of a current time step and a minimum weight value of the current time step, and the number of discrete weight states referring to the quantity of discrete weight states. The method also includes acquiring a weight state of a previous time step and a weight increment of the current time step and acquiring a state transfer direction by using a directional function according to the weight increment of the current time step. The method also includes acquiring a weight state of the current time step according to the weight state of the previous time step, the weight increment of the current time step, the state transfer direction, the weight value range and the number of discrete weight states. The method ensures that the weight value is always constrained in the same discrete-state space without the need for storing an additional virtual continuous-state implicit weight. On the premise that the computation performance of the neural network is ensured, the consumption of storage space is greatly reduced, and the computation complexity is reduced.

Neural network weight discretizing method, system, device, and readable storage medium

The present invention provides a multilevel read-only optical disk and the method for producing the same. The multilevel read-only optical disk comprises a plurality of the recording pits, each of the longitudinal sections of the recording pits has a multilevel arbitrary shape; the areas of the longitudinal sections of the recording pits are not identical each other; the area of each of the longitudinal sections of the recording pits is defined by the following formula: S=&#x222b;h(x)dx, where S represents the area of each of the longitudinal sections of the recording pits, x represents the coordinate value along the width direction of the recording pits, h(x) represents the pit-depth distribution function of the longitudinal sections of the recording pits and the region of the integration covers the whole longitudinal sections of the recording pits. Furthermore, the run length of the multilevel read-only optical disk is limited. The present invention obtains the advantage of both multilevel storage technology and run length limited (RLL) codes. They can improve prominently the storage capacity and data transfer rate of read-only optical disk without changing the wavelength of the laser and optical numerical aperture, and are most compatible with the conventional read-only optical disk system.

Multilevel read-only optical disk and method for producing the same

Growing community of inventors

Beijing, China

Jing Pei