Abstract
With the increase of glass detection speed, some defects of MapReduce distributed computing framework are exposed, and the processing speed and timeliness cannot meet the requirements of glass-defect detection in industrial technology. Based on the MapReduce distributed computing framework, this paper designs a threshold segmentation method to complete the segmentation of glass-defect images. By improving the replication placement strategy and pipeline scheduling mechanism, the computing and storage are localized, and the timeliness of data processing is accelerated. The experimental results show that the improved MapReduce computing framework has an average increase of 14.8% in processing speed. It can detect the glass ribbon running at 800m/h and also detect the number, position and type of defects on the glass ribbon.
Highlights
With the development of digital technology, digital image processing is widely used in industrial production
This paper implements the image threshold segmentation based on Hadoop framework by studying the application of distributed system to image processing
Experiments show that the data processing speed of improved MapReduce parallel computing framework is significantly improved, and the timeliness of the system is improved
Summary
With the development of digital technology, digital image processing is widely used in industrial production. The MapReduce framework was first used to process textual data collections, it has been recently used to process large images such as remote sensing images, high resolution images etc [3]. It is a data model based on key/value pairs. Each node intercepts the data in the corresponding interval (shuffle phase) [6] This process is the key for MapReduce to operate correctly, but it affects the speed of the system processing. The existing systems lack of research on improving the performance of MapReduce on large scale images processing. We take a large number of glass images which are collected online as test objects, realizing timely and accurate detection of various glass defects
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have