In order to generate an efficient common bitmap in single bitmap block truncation coding(SBBTC)of color images,an improved SBBTC scheme based on weighted plane(W-plane)method and hill climbing algorithm is proposed.Fi...In order to generate an efficient common bitmap in single bitmap block truncation coding(SBBTC)of color images,an improved SBBTC scheme based on weighted plane(W-plane)method and hill climbing algorithm is proposed.Firstly,the incoming color image is partitioned into non-overlapping blocks and each block is encoded using the W-plane method to get an initial common bitmap and quantization values.Then,the hill climbing algorithm is applied to optimize an initial common bitmap and generate a near-optimized common bitmap.Finally,the quantization values are recalculated by the near-optimized common bitmap and the considered color image is reconstructed block by block through the common bitmap and the new quantization values.Since the processing of each image block in SBBTC is independent and identical,parallel computing is applied to reduce the time consumption of this scheme.The simulation results show that the proposed scheme has better visual quality and time consumption than those of the reference SBBTC schemes.展开更多
为了挖掘满足用户特殊需求,如含指定项目数量的高效用项集(HUI),提出一种基于长度约束的蝙蝠高效用项集挖掘算法(HUIM-LC-BA)。该算法融合蝙蝠算法(BA)和长度约束构建高效用项集挖掘(HUIM)模型,首先将数据库转换为位图矩阵,实现高效的...为了挖掘满足用户特殊需求,如含指定项目数量的高效用项集(HUI),提出一种基于长度约束的蝙蝠高效用项集挖掘算法(HUIM-LC-BA)。该算法融合蝙蝠算法(BA)和长度约束构建高效用项集挖掘(HUIM)模型,首先将数据库转换为位图矩阵,实现高效的效用计算和数据库扫描;其次,采用重新定义的事务加权效用(RTWU)策略缩减搜索空间;最后,对项集进行长度修剪,使用深度优先搜索和轮盘赌注选择法确定修剪项目。在4个数据集的仿真实验中,当最大长度为6时,与HUIM-BA相比,HUIM-LC-BA挖掘的模式数量分别减少了91%、98%、99%与97%,同时运行时间也少于HUIM-BA;且在不同长度约束条件下,与FHM+(Faster High-utility itemset Ming plus)算法相比运行时间更稳定。实验结果表明,HUIM-LC-BA能有效挖掘具有长度约束的HUI,并减少挖掘模式的数量。展开更多
基金Supported by the National Natural Science Foundation of China(No.61402537)the Open Fund of Guangxi Key Laboratory of Hybrid Computation and IC Design Analysis(No.HCIC201706)the Sichuan Science and Technology Programme(No.2018GZDZX0041)
文摘In order to generate an efficient common bitmap in single bitmap block truncation coding(SBBTC)of color images,an improved SBBTC scheme based on weighted plane(W-plane)method and hill climbing algorithm is proposed.Firstly,the incoming color image is partitioned into non-overlapping blocks and each block is encoded using the W-plane method to get an initial common bitmap and quantization values.Then,the hill climbing algorithm is applied to optimize an initial common bitmap and generate a near-optimized common bitmap.Finally,the quantization values are recalculated by the near-optimized common bitmap and the considered color image is reconstructed block by block through the common bitmap and the new quantization values.Since the processing of each image block in SBBTC is independent and identical,parallel computing is applied to reduce the time consumption of this scheme.The simulation results show that the proposed scheme has better visual quality and time consumption than those of the reference SBBTC schemes.
文摘为了挖掘满足用户特殊需求,如含指定项目数量的高效用项集(HUI),提出一种基于长度约束的蝙蝠高效用项集挖掘算法(HUIM-LC-BA)。该算法融合蝙蝠算法(BA)和长度约束构建高效用项集挖掘(HUIM)模型,首先将数据库转换为位图矩阵,实现高效的效用计算和数据库扫描;其次,采用重新定义的事务加权效用(RTWU)策略缩减搜索空间;最后,对项集进行长度修剪,使用深度优先搜索和轮盘赌注选择法确定修剪项目。在4个数据集的仿真实验中,当最大长度为6时,与HUIM-BA相比,HUIM-LC-BA挖掘的模式数量分别减少了91%、98%、99%与97%,同时运行时间也少于HUIM-BA;且在不同长度约束条件下,与FHM+(Faster High-utility itemset Ming plus)算法相比运行时间更稳定。实验结果表明,HUIM-LC-BA能有效挖掘具有长度约束的HUI,并减少挖掘模式的数量。
