Such architectural improvements can improve the performance of traditional image edge detection algorithms under single-node architectures when processing massive datasets. Their results showed that the processing time for 8-bit images at a resolution of 1024 × 1024 was 122 ms, as well as a speedup ratio of approximately 5.39 times compared to the traditional Canny edge detection algorithm. Tang and Long proposed a fast implementation of the Canny operator based on a GPU+CPU combination in which the GPU was used to parallelize the Canny edge detection algorithm. With the advent of the big data era, traditional edge detection technologies are facing problem of poor edge detection and long runtimes thus, this technology needs to be further analyzed and studied. After nearly 60 years of research, many different edge detection methods have been designed, and each has its own characteristics and limitations. In 1959, Julesz was the first to discuss edge detection later, in 1965, Roberts began to systematically study edge detection. The performance of the edge detection algorithm directly affects the precision of extracted object contours and the performance of the system. Then, the edge strength of the image is defined, and the edge points are extracted by setting a threshold. First, an edge enhancement operator is used to highlight local edges in an image. Image edge detection algorithms have been widely studied.
In addition, edge detection is an important step in image analysis and 3D reconstruction and is therefore also an important feature in the field of digital image analysis. Edge detection is an important aspect of image processing and is the basis of many analytical methods in such fields as image segmentation, pattern recognition, machine vision, and regional shape extraction. Thus, edge detection technology is based on the discontinuity or mutation of gray levels or textural characteristics between an object and its background. Usually, an image edge is a set of pixels around which the gray-level values exhibit a step change. Most of an image's information is carried in the edges. The proposed algorithm in this study demonstrates both better edge detection performance and improved time performance.Įdges are a basic image feature and they are present between a target and a background and between two targets, two regions, or two primitives. Overall, our approach system speeds up the system by approximately 3.4 times when processing large-scale datasets, which demonstrates the obvious superiority of our method.
The parallel approach reduced the running time by approximately 67.2% on a Hadoop cluster architecture consisting of 5 nodes with a dataset of 60,000 images. The proposed parallel Otsu-Canny edge detection algorithm performs better than other traditional edge detection algorithms.
For the experiments, we constructed datasets of different scales from the Pascal VOC2012 image database. The Otsu algorithm is used to optimize the Canny operator's dual threshold and improve the edge detection performance, while the MapReduce parallel programming model facilitates parallel processing for the Canny operator to solve the processing speed and communication cost problems that occur when the Canny edge detection algorithm is applied to big data. To improve the runtime and edge detection performance of the Canny operator, in this paper, we propose a parallel design and implementation for an Otsu-optimized Canny operator using a MapReduce parallel programming model that runs on the Hadoop platform. However, as the size of the image dataset increases, the edge detection performance of the Canny operator decreases and its runtime becomes excessive. The Canny operator is widely used to detect edges in images.
0 kommentar(er)
