Near-sensor visual perception processing chip and Internet of Things sensing device

Abstract

The embodiment of the invention provides a near-sensor visual perception processing chip and an Internet of Things sensing device, wherein the chip comprises a control unit and an analog processing unit, the control unit is used for inputting the binarized analog data into the analog processing unit, the binarized analog data is composed of an analog voltage signal acquired by a sensor and a weight signal corresponding to the analog voltage signal, and the analog processing unit is used for processing the received binary analog data to obtain a current value corresponding to the analog voltage signal. The sensing device comprises a CMOS image sensor, the chip and a communication module. According to the near-sensor visual perception processing chip and the Internet of Things sensing device provided by the embodiment of the present invention, the analog processing unit adopting a binary neural network algorithm is placed behind the CMOS image sensor and in front of the ADC, so that the chip can directly process the analog voltage signals, the huge energy consumption during the analog-to-digital conversion is avoided, and the energy efficiency is effectively improved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a proximity sensor visual perception processing chip and an Internet of Things sensing device. Background technique [0002] With the rapid development of artificial intelligence, the current Intelligent Internet of Things (abbreviation: AIoT) has been widely used in our production and life. However, AIoT devices not only perceive image data at the terminal node, especially in the image acquisition and processing stage, but also need to analyze the acquired image data. However, due to the limitation of battery resources, especially when dealing with computationally intensive tasks applying artificial intelligence algorithms, various AIoT devices often cannot strike a balance between processing quality, real-time performance, power or energy constraints. [0003] In order to solve the above existing problems and improve the processing efficiency of AIoT devices in the fi...

AI Technical Summary

Problems solved by technology

However, due to the limitation of battery resources, especially when dealing with computationally intensive tasks applying artificial intelligence algorithms, it is often impossible for various AIoT devices to strike a balance between processing quality, real-time performance, power or energy constraints

[0003] In order to solve the above existing problems and improve the processing efficiency of AIoT devices in the field of edge computing, at present, the following methods are mainly used to realize the acquisition and processing of image data by AIoT device terminals: First, by using embedded computer graphics processing processor (GPU) or based on field-programmable gate series (FPGA), but this solution still has high power consumption and cannot actually solve the problem of battery resource limitations

The second is that the custom neural network processing unit (NPU) and the high-speed custom machine learning chip (TPU) optimized on its architecture are used in the visual perception unit of the AIoT device, but the application of the above-mentioned digital signal processing scheme Technically, digital NPU-based accelerators still suffer from low-energy-efficiency ADC conversion, especially when sensing acquisition and processing high-resolution data, which will greatly reduce the performance of the sensing system

Images