Camera system and image recognition system

Abstract

The invention discloses a camera system and an image identification system. The camera system comprises a cell array calculation system and an image sensor; the cell array calculation system comprises a main control CPU, a cell array and a cell array bus; the cell array is a two-dimension array which is composed of more than one cell having functions of calculation and storage; the image sensor is a two-dimension array composing of more than one image collection units; the cell array is stacked with the image sensor to form a three-dimension structure; the image collection unit in the image sensor is correspondingly connected to the cell in the cell array on one-to-one basis; the image collection unit collects the image data for the cell in the cell array to process; the cell array calculation system comprises at least one of a storage unit array and a memory unit array; the main control CPU performs communication with each cell in the cell array through the cell array bus; and communication interfaces are arranged between the adjacent cells for mutually transmitting data. The invention enables the image processing capability of the camera system to dramatically improve.

Description

technical field [0001] The invention relates to the technical field of computers and computer applications, in particular to a camera system and an image recognition system. Background technique [0002] Generally speaking, a computer mainly includes three core parts: central processing unit (CPU, Central Processing Unit), memory and storage. [0003] Through the tireless efforts of some of the world's top companies, CPUs have evolved into extremely complex semiconductor chips. The number of MOS tubes inside the top-level CPU core can exceed 100 million. The current industry trend is subject to power consumption, and it is difficult to increase the operating frequency of the CPU. It is also difficult to improve the operating efficiency of modern CPUs, which are already extremely complex. New CPU products are increasingly evolving towards multi-core. [0004] In terms of memory, dynamic random access memory (DRAM, Dynamic RandomAccess Memory) technology is currently domin...

AI Technical Summary

Problems solved by technology

[0009] High-end image sensors can obtain image data in a very short period of time (microsecond level), but the data volume of a frame of high-definition image is very large. In the current camera system, considering the existing computer architecture due to CPU and memory 1. The communication bottleneck between the storages affects the overall performance of the computer. It usually takes 1 / 30-1 / 60 seconds to read out the image data. Therefore, in most camera systems, the video processing capability Can't keep up with the speed of getting image data

However, the writing speed of traditional hard disks, solid-state hard disks, and flash memory is very slow, and the data volume of high-speed cameras is large, and it is difficult for traditional storage devices to keep up

Images