An artificial intelligence scoring system based on edge computing

Abstract

The present invention disclosed an artificial intelligence scroll system based on edge -based computing, including: image collection device and local terminal; local terminals connect to image collection devices, local terminals include control modules, local terminals include image processing modules, image processing modulesIncluding: information recognition unit, image segmentation unit, and computing unit; local terminals also include summary modules, and the summary module is connected to the computing unit. The local terminal also includes the detection module and display screen, and the detection module and the display are connected.The invention deployed the embedded neural network processing chip with strong computing ability on the local terminal, which improved the computing power of the local terminal, set the calculation volume of the scroll on the local terminal, release the computing pressure on the cloud server, saved resources resourcesIt can be suitable for large -scale promotion and use.

Description

Technical field [0001] The present invention involves the field of intelligent scrolls and is more specific to involve a marginal -based artificial intelligence scroll system. Background technique [0002] With the rapid development of computers and artificial intelligence, it has gradually been applied to many fields of work and life. Among them, in the process of education and teaching, the examination is an effective evaluation method of teaching results. After a test is completed The analysis work will occupy a lot of time for teachers, and even compressed teachers' lesson preparation time, so the related technologies of computer intelligent scrolls have appeared. The current method is usually used to combine the computer and the scanner. After scanning the test papers in batches, upload the scanned test paper image to the cloud, and perform algorithms on the image in the cloud to get the corresponding test paper (including including the corresponding test paper (including in...

AI Technical Summary

Problems solved by technology

[0003] However, there are still some drawbacks to the above-mentioned current marking methods. For example, the examination time of most domestic schools is relatively concentrated, and a large amount of concurrency will be generated during the concentrated examination time, that is, there are a large number of examination paper images in a short period of time. Uploading to the cloud requires computing and processing on the cloud. At this time, the computing load on the cloud is heavy, and the computing requirements on the cloud are relatively high. As a result, this marking method is not conducive to large-scale promotion across the country; in the above scheme, the scanned test paper The image needs to be uploaded to the cloud first. If there is any abnormality after cloud processing, it needs to be sent to the computer for manual confirmation. Therefore, the entire marking process needs to be on-site, and when the network speed is slow, the test paper will be uploaded. Slow, resulting in extremely low efficiency of the entire marking process; in addition, in different schools, the school's computer configuration is not the same, resulting in large-scale promotion of applications, often in the actual scene compatibility problems, causing the entire marking process to freeze Or crash, affecting the user experience

Images