Disjoint sampling is critical for rigorous and unbiased evaluation of state-of-the-art(SOTA)models e.g.,Attention Graph and Vision Transformer.When training,validation,and test sets overlap or share data,it introduces...Disjoint sampling is critical for rigorous and unbiased evaluation of state-of-the-art(SOTA)models e.g.,Attention Graph and Vision Transformer.When training,validation,and test sets overlap or share data,it introduces a bias that inflates performance metrics and prevents accurate assessment of a model’s true ability to generalize to new examples.This paper presents an innovative disjoint sampling approach for training SOTA models for the Hyperspectral Image Classification(HSIC).By separating training,validation,and test data without overlap,the proposed method facilitates a fairer evaluation of how well a model can classify pixels it was not exposed to during training or validation.Experiments demonstrate the approach significantly improves a model’s generalization compared to alternatives that include training and validation data in test data(A trivial approach involves testing the model on the entire Hyperspectral dataset to generate the ground truth maps.This approach produces higher accuracy but ultimately results in low generalization performance).Disjoint sampling eliminates data leakage between sets and provides reliable metrics for benchmarking progress in HSIC.Disjoint sampling is critical for advancing SOTA models and their real-world application to large-scale land mapping with Hyperspectral sensors.Overall,with the disjoint test set,the performance of the deep models achieves 96.36%accuracy on Indian Pines data,99.73%on Pavia University data,98.29%on University of Houston data,99.43%on Botswana data,and 99.88%on Salinas data.展开更多
针对景区手写诗词存在背景纹理复杂、字体尺寸及风格多样等特点导致景区游客难以识别手写诗词的问题,首先,分析研究景区手写诗词的识别场景,设计景区诗词检测网络(detection of poetry in scenic areas-network,DPSA-Net)以提取景区手...针对景区手写诗词存在背景纹理复杂、字体尺寸及风格多样等特点导致景区游客难以识别手写诗词的问题,首先,分析研究景区手写诗词的识别场景,设计景区诗词检测网络(detection of poetry in scenic areas-network,DPSA-Net)以提取景区手写诗词不同尺度的特征,并结合手写诗词字符间的链接依赖关系实现景区手写诗词检测;其次,设计了卷积循环聚合网络(convolution recurrent aggregation network,CRA-Net)以对景区手写诗词进行识别,结合卷积神经网络(convolutional neural networks,CNN)和双向长短期记忆网络提取手写诗词图像的序列特征,并通过聚合交叉熵(aggregation cross-entropy,ACE)实现特征向文本的转换;最后,结合景区知识图谱对CRA-Net的输出进行校正,进而提高景区手写诗词的识别准确率。实验结果表明,通过景区手写诗词矫正技术对CRA-Net的识别结果矫正后,识别准确率达到了79.04%,同时,该技术具有较好的抗干扰能力和良好的应用前景。展开更多
在数字芯片设计后端流程中,宏和标准单元的布局是一项耗时的工作,通过机器学习快速有效地提供解决方案能够加快芯片开发的周期,降低人工布局带来的风险;然而布局问题是一个多目标优化问题,目前大多数方法都注重在满足各项指标下最大化...在数字芯片设计后端流程中,宏和标准单元的布局是一项耗时的工作,通过机器学习快速有效地提供解决方案能够加快芯片开发的周期,降低人工布局带来的风险;然而布局问题是一个多目标优化问题,目前大多数方法都注重在满足各项指标下最大化减小线长,已换取时钟延迟的降低,忽略了其他指标仍然存在下降的空间,例如良好的拥塞指标有利于降低芯片散热和功耗;针对上述问题,设计一种新的带有密集型奖励函数的深度强化学习框架,将拥塞信息映射到图像中,给出新的特征嵌入模型对版图的全局信息进行多尺度提取,并引入图注意力网络捕获网表的连接关系,采用Advantage Actor Critic(A2C)算法更新策略函数,实现了数字版图的自动布局,并在公共的数字芯片网表基准上验证了该方法的有效性。展开更多
传统视觉即时定位与建图(SLAM)算法若无回环检测可能会存在累积误差无法消除的现象,即使有回环检测,也因准确率和效率比较低而无法应用于轻量级设备上,为此,研究一种回环检测优化的视觉SLAM算法.前端估计时,对相邻帧图像进行ORB(oriente...传统视觉即时定位与建图(SLAM)算法若无回环检测可能会存在累积误差无法消除的现象,即使有回环检测,也因准确率和效率比较低而无法应用于轻量级设备上,为此,研究一种回环检测优化的视觉SLAM算法.前端估计时,对相邻帧图像进行ORB(oriented fast and rotated brief)特征提取与匹配,对匹配成功的特征点进行PnP(perspective-n-point)求解,获得相机运动估计并筛选出关键帧图像;后端优化时,利用SqueezeNet卷积神经网络(CNN)提取图像的特征向量,计算余弦相似度判断是否出现回环,若出现回环则在位姿图中增加相应约束,利用图优化理论对全局位姿进行整体优化;最后利用项目组制作的数据集和TUM(technical university of munich)公开数据集进行测试与对比.研究结果表明:相比于无回环检测算法,本文方法可以成功检测到回环并为全局轨迹优化增添约束;相比于传统词袋法,在回环检测准确率相同的情况下,本文方法召回率可提高21%且计算耗时减少74%;与RGB-D(red green blue-depth)SLAM算法相比,本文方法建图误差可降低29%.展开更多
基金the Researchers Supporting Project number(RSPD2024R848),King Saud University,Riyadh,Saudi Arabia.
文摘Disjoint sampling is critical for rigorous and unbiased evaluation of state-of-the-art(SOTA)models e.g.,Attention Graph and Vision Transformer.When training,validation,and test sets overlap or share data,it introduces a bias that inflates performance metrics and prevents accurate assessment of a model’s true ability to generalize to new examples.This paper presents an innovative disjoint sampling approach for training SOTA models for the Hyperspectral Image Classification(HSIC).By separating training,validation,and test data without overlap,the proposed method facilitates a fairer evaluation of how well a model can classify pixels it was not exposed to during training or validation.Experiments demonstrate the approach significantly improves a model’s generalization compared to alternatives that include training and validation data in test data(A trivial approach involves testing the model on the entire Hyperspectral dataset to generate the ground truth maps.This approach produces higher accuracy but ultimately results in low generalization performance).Disjoint sampling eliminates data leakage between sets and provides reliable metrics for benchmarking progress in HSIC.Disjoint sampling is critical for advancing SOTA models and their real-world application to large-scale land mapping with Hyperspectral sensors.Overall,with the disjoint test set,the performance of the deep models achieves 96.36%accuracy on Indian Pines data,99.73%on Pavia University data,98.29%on University of Houston data,99.43%on Botswana data,and 99.88%on Salinas data.
文摘在数字芯片设计后端流程中,宏和标准单元的布局是一项耗时的工作,通过机器学习快速有效地提供解决方案能够加快芯片开发的周期,降低人工布局带来的风险;然而布局问题是一个多目标优化问题,目前大多数方法都注重在满足各项指标下最大化减小线长,已换取时钟延迟的降低,忽略了其他指标仍然存在下降的空间,例如良好的拥塞指标有利于降低芯片散热和功耗;针对上述问题,设计一种新的带有密集型奖励函数的深度强化学习框架,将拥塞信息映射到图像中,给出新的特征嵌入模型对版图的全局信息进行多尺度提取,并引入图注意力网络捕获网表的连接关系,采用Advantage Actor Critic(A2C)算法更新策略函数,实现了数字版图的自动布局,并在公共的数字芯片网表基准上验证了该方法的有效性。
文摘传统视觉即时定位与建图(SLAM)算法若无回环检测可能会存在累积误差无法消除的现象,即使有回环检测,也因准确率和效率比较低而无法应用于轻量级设备上,为此,研究一种回环检测优化的视觉SLAM算法.前端估计时,对相邻帧图像进行ORB(oriented fast and rotated brief)特征提取与匹配,对匹配成功的特征点进行PnP(perspective-n-point)求解,获得相机运动估计并筛选出关键帧图像;后端优化时,利用SqueezeNet卷积神经网络(CNN)提取图像的特征向量,计算余弦相似度判断是否出现回环,若出现回环则在位姿图中增加相应约束,利用图优化理论对全局位姿进行整体优化;最后利用项目组制作的数据集和TUM(technical university of munich)公开数据集进行测试与对比.研究结果表明:相比于无回环检测算法,本文方法可以成功检测到回环并为全局轨迹优化增添约束;相比于传统词袋法,在回环检测准确率相同的情况下,本文方法召回率可提高21%且计算耗时减少74%;与RGB-D(red green blue-depth)SLAM算法相比,本文方法建图误差可降低29%.
