ID 原文 译文
983 现有研究多基于单特征且未挖掘属性蕴含的深层语义，故无法准确刻画图像内容。 However, a single feature is used to characterize images and the deep-level semantics are notfully explored. So annotations cannot depict images comprehensively.
984 改进有效区域基因选择算法融合图像特征， The traditional effective range based gene selection al-gorithm is modified to complete feature fusion.
985 并设计迁移学习策略，实现材质属性标注; And transfer learning strategy is designed to complete material annotation.
986 基于判别相关分析挖掘特征间跨模态语义， The cross-modal semantics among features are mined by the discriminant correlation analysis algorithm.
987 以改进相对属性模型，标注材质属性蕴含的深层语义-实用属性。 So the relative at-tribute model is optimized to complete deep-level semantics (practical attributes)annotation.
988 实验表明:材质属性标注精准度达 63.11% ，较最强基线提升 1.97% ; Experimental results demon-strate:Material attributes annotation accuracy reaches 63. 11% , which is improved by 1. 97% compared with baseline.
989 实用属性标注精准度达59.15% ，较最强基线提升 2.85% ; Practi-cal attributes annotation accuracy reaches 59. 15% , which is improved by 2. 85% compared with baseline.
990 层次化的标注结果能全面刻画图像内容。 The proposed hi-erarchical annotation mechanism can more comprehensively depict images.
991 通过增加伪码字的代价，基于交替方向乘子法(Alternating Direction Method of Multipliers，ADMM)的惩罚译码方法可以改善低密度奇偶校验(Low-Density Parity-Check，LDPC)码低信噪比区域的译码性能，同时具有低的译码复杂度。 By making the pseudocodewords more costly, the penalized decoding method based on alternating directionmethod of multipliers (ADMM)can improve the decoding performance for low-density parity-check (LDPC)codes at lowsignal-to-noise ratios and also has low decoding complexity.
992 而减少 ADMM 惩罚译码的欧几里德投影次数、选择合适的消息调度策略和设计有效的罚函数是提高ADMM 惩罚译码速度的三种重要方法。 Reducing the number of Euclidean projection in ADMM penal-ized decoding, selecting the appropriate message scheduling strategy and designing effective penalty function are three impor-tant methods to increase the ADMM penalized decoding speed.