Scholarlink

<p>The technique of breaking down photos into distinct sections is known as image segmentation. This work aims to divide the images into sub-images to make the image analysis easier. The traditional techniques consume more time to compute edge and boundary information, creating computation difficulties, especially when the images contain noisy pixels. In addition, the boundary and corner detection process faces optimization problems. The research issues are overcome by applying the Slap-Optimized Sequence Deep Neural Networks (SO-SDNN). Initially, the input images are captured and processed using the median filter, which removes noise by examining each pixel. Then, the sequence deep neural model explores each pixel to explore the boundary and edge information. During the analysis, the slap optimization model utilizes the pixel position and distribution to minimize the optimization problem. By optimizing the network parameters, the optimization method maximizes the overall effectiveness of edge detection. SO-SDNN focuses on effectively calculating edge and boundary information, dealing with noisy pictures, optimizing boundary and corner detection, and optimizing network parameters to enhance total efficiency of edge detection in image segmentation tasks. Efficiency of system was evaluated using respective performance metrics, and the method ensures a high detection rate of 98.105% for identifying the edges, exceeding the performance of alternative approaches.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep neural network-based image segmentation system to create slap-optimized sequence

  • 期刊论文

  • 发表日期：

    2025年10月11日

查看全文
  • M. Jansirani,
  • P. Sumitra

摘要

The technique of breaking down photos into distinct sections is known as image segmentation. This work aims to divide the images into sub-images to make the image analysis easier. The traditional techniques consume more time to compute edge and boundary information, creating computation difficulties, especially when the images contain noisy pixels. In addition, the boundary and corner detection process faces optimization problems. The research issues are overcome by applying the Slap-Optimized Sequence Deep Neural Networks (SO-SDNN). Initially, the input images are captured and processed using the median filter, which removes noise by examining each pixel. Then, the sequence deep neural model explores each pixel to explore the boundary and edge information. During the analysis, the slap optimization model utilizes the pixel position and distribution to minimize the optimization problem. By optimizing the network parameters, the optimization method maximizes the overall effectiveness of edge detection. SO-SDNN focuses on effectively calculating edge and boundary information, dealing with noisy pictures, optimizing boundary and corner detection, and optimizing network parameters to enhance total efficiency of edge detection in image segmentation tasks. Efficiency of system was evaluated using respective performance metrics, and the method ensures a high detection rate of 98.105% for identifying the edges, exceeding the performance of alternative approaches.