The proceedings contain 15 papers from the conference on parallel and distributed methods for image processing II. The topics discussed include: parallel DSP with memory and I/O processors;analog VLSI implementation o...
详细信息
The proceedings contain 15 papers from the conference on parallel and distributed methods for image processing II. The topics discussed include: parallel DSP with memory and I/O processors;analog VLSI implementation of a morphological associative memory;real-time parallel video imageprocessing on a PC cluster;thread concept for automatic task parallelization in image analysis;new parallel vision environment in heterogeneous networked computing and toolkit for parallelimageprocessing.
Advances in the imageprocessing field have brought new methods which are able to perform complex tasks robustly. However, in order to meet constraints on functionality and reliability, imaging application developers ...
详细信息
ISBN:
(纸本)9780819484093
Advances in the imageprocessing field have brought new methods which are able to perform complex tasks robustly. However, in order to meet constraints on functionality and reliability, imaging application developers often design complex algorithms with many parameters which must be finely tuned for each particular environment. The best approach for tuning these algorithms is to use an automatic training method, but the computational cost of this kind of training method is prohibitive, making it inviable even in powerful machines. The same problem arises when designing testing procedures. This work presents methods to train and test complex imageprocessing algorithms in parallel execution environments. The approach proposed in this work is to use existing resources in offices or laboratories, rather than expensive clusters. These resources are typically non-dedicated, heterogeneous and unreliable. The proposed methods have been designed to deal with all these issues. Two methods are proposed: intelligent training based on genetic algorithms and PVM, and a full factorial design based on grid computing which can be used for training or testing. These methods are capable of harnessing the available computational power resources, giving more work to more powerful machines, while taking its unreliable nature into account. Both methods have been tested using real applications.
Structure from Motion (SfM) is a fundamental computer vision technique that recovers scene structure and camera motion from multi-view images. When facing large-scale scenarios, cluster-based methods are commonly empl...
详细信息
ISBN:
(纸本)9789819785070;9789819785087
Structure from Motion (SfM) is a fundamental computer vision technique that recovers scene structure and camera motion from multi-view images. When facing large-scale scenarios, cluster-based methods are commonly employed to improve reconstruction efficiency. However, these methods currently face challenges regarding their limited robustness, redundant computation, and drift. To address these issues, we propose a unified pipeline called ER-SfM, which enhances the three key aspects of cluster-based SfM: image clustering, local reconstruction, and merging. In terms of image clustering, we propose a three-stage image clustering method to ensure adequate and reliable connections between clusters. In the local reconstruction stage, we expedite the reconstruction process by eliminating duplicate point cloud computation. In the final merging stage, we introduce a global merging algorithm without scale ambiguity to address the drift problem. Extensive experimental results demonstrate the superior performance of our method in terms of both robustness and efficiency compared to state-of-the-art methods.
Accurate and rapid classification of large-scale lychee images is crucial for collecting germplasm resources and studying the characteristics of different lychee varieties, and it requires the construction of accurate...
详细信息
Accurate and rapid classification of large-scale lychee images is crucial for collecting germplasm resources and studying the characteristics of different lychee varieties, and it requires the construction of accurate classification models and the design of rapid classification algorithms. However, the current deep learning-based classification methods for lychee images are unable to simultaneously meet the processing requirements of accuracy and timeliness in large-scale lychee image classification. To address the problem above, this paper proposes a largescale parallel classification algorithm for lychee images based on Spark and deep learning. Specifically, first, the T_ECBAM_ResNetS-34 model architecture was designed and trained using a self-built dataset covering ten types of lychee images and the PyTorch deep learning framework, which improved the accuracy of model classification;Second, the model inference algorithm trained by PyTorch was restructured, utilizing Apache Spark RDD and broadcast variables and data structures to implement data partitioning and model parallel computation across nodes. The experimental results show that the method proposed in this paper surpasses existing technologies in both classification accuracy and the speed of large-scale lychee image classification.
Multi-image optical encryption (MOE) has demonstrated promising potential in image data protection owing to its parallelprocessing capability and abundant degrees of freedom. However, existing methods suffer from eit...
详细信息
Multi-image optical encryption (MOE) has demonstrated promising potential in image data protection owing to its parallelprocessing capability and abundant degrees of freedom. However, existing methods suffer from either low compression ratios or stringent experimental conditions, such as accurate calibration of phase modulation, precise manufacturing of encryption elements, and no ambient light interference. This work introduces a lensless sparse point spread function-based multi-image optical encryption (sPSF-MOE) technique that addresses these challenges and enhances performance. In the encryption process, each plaintext image is encoded using a sparsely distributed PSF with specifically designed geometric shapes through spatial phase engineering. The resulting ciphertexts are superimposed to produce a compressed ciphertext. During decryption, an iterative algorithm recovers encrypted images with improved reconstruction quality. We show that sPSF-MOE ensures high fidelity for binary (gray-scale) images at a compression ratio of 12 (6) and resists autocorrelation-based attacks. Integrating principal component analysis (PCA) into decryption preserves image high fidelity under ambient light interference. sPSF-MOE reduces the bandwidth requirement for data transmission while ensuring data integrity.
Visual Question Answering (VQA) is a challenging task that bridges the computer vision and natural language processing communities. It provide natural language answers to questions related to an associated image. Most...
详细信息
Visual Question Answering (VQA) is a challenging task that bridges the computer vision and natural language processing communities. It provide natural language answers to questions related to an associated image. Most existing VQA methods focus on the fusion and inference of visual features with the textual question. However, visual features often lack the necessary semantic information required to answer the questions accurately. To address this limitation, we propose a novel approach called Question-Guided parallel Attention (QGPA), which effectively leverages the semantic information provided by an embedded image captioning model to answer related questions. First, we introduce an Attention-Aware (AA) mechanism that extends the traditional attention mechanism, helping to filter out incorrect or irrelevant information during answer prediction. Second, QGPA incorporates AA, which simultaneously utilizes visual features and semantic information from the embedded image captioning model to answer questions. Experiments results demonstrate that the accuracy of "Overall" of our proposed model delivers 72.57% and 72.55% on the test-dev and test-std split set of VQA-v2.0 dataset, respectively, which outperforms most existing VQA methods. The experiment results and ablation studies demonstrate that the proposed method has good performance.
Adversarial attacks are now becoming quite a dangerous means of disrupting imageprocessing systems that use machine learning methods for decision making. Therefore, developing effective countermeasures against advers...
详细信息
暂无评论