results: 提高图像质量并减少后处理步骤，可以进一步发展为多种任务专业无镜头成像系统Abstract
Lensless imagers based on diffusers or encoding masks enable high-dimensional imaging from a single shot measurement and have been applied in various applications. However, to further extract image information such as edge detection, conventional post-processing filtering operations are needed after the reconstruction of the original object images in the diffuser imaging systems. Here, we present the concept of a temporal compressive edge detection method based on a lensless diffuser camera, which can directly recover a time sequence of edge images of a moving object from a single-shot measurement, without further post-processing steps. Our approach provides higher image quality during edge detection, compared with the conventional post-processing method. We demonstrate the effectiveness of this approach by both numerical simulation and experiments. The proof-of-concept approach can be further developed with other image post-process operations or versatile computer vision assignments toward task-oriented intelligent lensless imaging systems.
摘要
Diffuser 或编码面增强的镜头less imaging 技术可以实现高维度图像取得，并在不同应用中使用。然而，为了进一步提取图像信息，例如边检测，传统的后处理滤波操作是必要的。在这种情况下，我们提出了基于减压 diffuser 摄像机的时间压缩边检测方法，可以从单个测量中直接回收移动物体的时间序列边图像，无需进一步的后处理步骤。我们的方法可以提供更高的图像质量，比传统的后处理方法更高。我们通过数值模拟和实验证明了这种方法的有效性。这种概念可以进一步发展为任务导向的智能镜头less imaging 系统。
Improving HEVC Encoding of Rendered Video Data Using True Motion Information
paper_authors: Christian Herglotz, David Müller, Andreas Weinlich, Frank Bauer, Michael Ortner, Marc Stamminger, André Kaup
for: 提高计算机生成视频序列的编码过程
methods: 利用计算机生成的动态vector进行增强评估性能
results: 实现了3.78%的均值bitrate减少Abstract
This paper shows that motion vectors representing the true motion of an object in a scene can be exploited to improve the encoding process of computer generated video sequences. Therefore, a set of sequences is presented for which the true motion vectors of the corresponding objects were generated on a per-pixel basis during the rendering process. In addition to conventional motion estimation methods, it is proposed to exploit the computer generated motion vectors to enhance the ratedistortion performance. To this end, a motion vector mapping method including disocclusion handling is presented. It is shown that mean rate savings of 3.78% can be achieved.
摘要
这个论文表明，可以使用场景中对象的真实运动向量来改善计算机生成视频序列的编码过程。因此，一个集合的序列被提供，其中对应的对象的真实运动向量在渲染过程中按照每个像素基础生成。除了传统的运动估计方法外，还提议使用计算机生成的运动向量来提高评估性能。为此，一种运动向量映射方法，包括缺失处理，被提出。实验表明，可以获得3.78%的均值Rate savings。
Deep Learning-based Synthetic High-Resolution In-Depth Imaging Using an Attachable Dual-element Endoscopic Ultrasound Probe
paper_authors: Hah Min Lew, Jae Seong Kim, Moon Hwan Lee, Jaegeun Park, Sangyeon Youn, Hee Man Kim, Jihun Kim, Jae Youn Hwang for: This paper aims to provide clinicians with appropriate hardware specifications for precise diagnosis by enhancing the resolution and penetration depth of endoscopic ultrasound (EUS) imaging.methods: The proposed approach uses a novel deep learning-based high-resolution in-depth imaging probe that offers low- and high-frequency ultrasound image pairs. The probe is designed with customized low- and high-frequency ultrasound transducers and a special geared structure to enable the same image plane.results: The proposed approach was evaluated with a wire phantom and a tissue-mimicking phantom, and 442 ultrasound image pairs were acquired from the tissue-mimicking phantom. The results demonstrate the feasibility of the approach for providing synthetic high-resolution in-depth images deep inside tissues, and a suitable deep-learning model was identified for the task.Abstract
Endoscopic ultrasound (EUS) imaging has a trade-off between resolution and penetration depth. By considering the in-vivo characteristics of human organs, it is necessary to provide clinicians with appropriate hardware specifications for precise diagnosis. Recently, super-resolution (SR) ultrasound imaging studies, including the SR task in deep learning fields, have been reported for enhancing ultrasound images. However, most of those studies did not consider ultrasound imaging natures, but rather they were conventional SR techniques based on downsampling of ultrasound images. In this study, we propose a novel deep learning-based high-resolution in-depth imaging probe capable of offering low- and high-frequency ultrasound image pairs. We developed an attachable dual-element EUS probe with customized low- and high-frequency ultrasound transducers under small hardware constraints. We also designed a special geared structure to enable the same image plane. The proposed system was evaluated with a wire phantom and a tissue-mimicking phantom. After the evaluation, 442 ultrasound image pairs from the tissue-mimicking phantom were acquired. We then applied several deep learning models to obtain synthetic high-resolution in-depth images, thus demonstrating the feasibility of our approach for clinical unmet needs. Furthermore, we quantitatively and qualitatively analyzed the results to find a suitable deep-learning model for our task. The obtained results demonstrate that our proposed dual-element EUS probe with an image-to-image translation network has the potential to provide synthetic high-frequency ultrasound images deep inside tissues.
摘要
挺投射 ultrasound (EUS) 的分辨率和扩散深度之间存在负权衡。在考虑人体内部器官的实际特点下，为临床诊断提供合适的硬件参数是必要的。目前，使用深度学习技术进行超解像 (SR) ultrasound 图像修复的研究已经很多，但大多数这些研究并没有考虑 ultrasound 图像的特点，而是基于下采样的 ultrasound 图像进行 SR 技术。本研究提出了一种基于深度学习的高解像深度扫描仪，能够提供低频和高频 ultrasound 图像对。我们开发了一种可附加的双元 EUS 探针，其中包括特制的低频和高频 ultrasound 传感器。我们还设计了一种特殊的几何结构，以使得同一个图像平面。我们对用一束织物和一种组织模拟物进行评估。然后，我们获得了442个 ultrasound 图像对。我们应用了多种深度学习模型，以获得 sintetic 高解像深度图像，从而证明了我们的方法的可行性。此外，我们对结果进行了量化和质量分析，以选择适合我们任务的深度学习模型。获得的结果表明，我们的提议的双元 EUS 探针与嵌入式深度学习网络具有将高频 ultrasound 图像深入到组织中的潜力。
Improving Deep Learning-based Defect Detection on Window Frames with Image Processing Strategies
paper_authors: Jorge Vasquez, Hemant K. Sharma, Tomotake Furuhata, Kenji Shimada
for: The paper is written for researchers and manufacturers who are interested in using machine learning and computer vision techniques for defect detection in window frames, particularly in challenging environments like construction sites.
methods: The paper proposes a novel defect detection pipeline called InspectNet, which combines image enhancement and augmentation techniques with a pre-trained U-Net model for window frame defect detection and segmentation. The pipeline is designed to improve the accuracy of defect detection in challenging environments.
results: The paper presents the results of experiments conducted using a Spot Robot for window frame inspections, with 16 variations of the dataset constructed using different image augmentation settings. The results show that the proposed InspectNet pipeline outperformed other algorithms when image enhancement and augmentation techniques were applied, achieving an average Intersection over Union (IoU) value of 0.91 when using the best dataset.Abstract
Detecting subtle defects in window frames, including dents and scratches, is vital for upholding product integrity and sustaining a positive brand perception. Conventional machine vision systems often struggle to identify these defects in challenging environments like construction sites. In contrast, modern vision systems leveraging machine and deep learning (DL) are emerging as potent tools, particularly for cosmetic inspections. However, the promise of DL is yet to be fully realized. A few manufacturers have established a clear strategy for AI integration in quality inspection, hindered mainly by issues like scarce clean datasets and environmental changes that compromise model accuracy. Addressing these challenges, our study presents an innovative approach that amplifies defect detection in DL models, even with constrained data resources. The paper proposes a new defect detection pipeline called InspectNet (IPT-enhanced UNET) that includes the best combination of image enhancement and augmentation techniques for pre-processing the dataset and a Unet model tuned for window frame defect detection and segmentation. Experiments were carried out using a Spot Robot doing window frame inspections . 16 variations of the dataset were constructed using different image augmentation settings. Results of the experiments revealed that, on average, across all proposed evaluation measures, Unet outperformed all other algorithms when IPT-enhanced augmentations were applied. In particular, when using the best dataset, the average Intersection over Union (IoU) values achieved were IPT-enhanced Unet, reaching 0.91 of mIoU.
摘要
检测窗框中微型瑕疵，包括折叠和擦抹，是维护产品完整性和保持品牌形象的关键。传统的机器视觉系统经常在建筑现场中难以检测这些瑕疵。相比之下，现代视觉系统通过机器学习和深度学习（DL）在cosmetic检测方面表现出了强大的潜力。然而，DL的承诺仍未实现。一些制造商在质检中采用了明确的AI整合策略，主要受到数据质量缺乏和环境变化所迟缓。为了解决这些挑战，本研究提出了一种创新的瑕疵检测方法，包括适应数据资源的限制的图像增强和扩展技术，以及适应窗框瑕疵检测和分 segmentation的Unet模型。实验使用了一个Spot Robot进行窗框检测。根据不同的图像扩展设置，建立了16个变种的数据集。实验结果表明，在所有提议的评价指标中，Unet模型在IPT-加强的扩展设置下表现最佳，特别是在使用最佳数据集时，Unet模型的平均交集率（IoU）值达0.91。