A5011631699- Sparse and Compressive Sensing Techniques
- CCD and CMOS Imaging Sensors
- Advanced Data Compression Techniques
- Blind Source Separation Techniques
- Photoacoustic and Ultrasonic Imaging
- Analog and Mixed-Signal Circuit Design
- Advanced Image and Video Retrieval Techniques
- Digital Image Processing Techniques
- Video Coding and Compression Technologies
- Image and Video Quality Assessment
- Generative Adversarial Networks and Image Synthesis
- Forecasting Techniques and Applications
- Image Processing Techniques and Applications
- Digital Media and Visual Art
- Domain Adaptation and Few-Shot Learning
- Stock Market Forecasting Methods
- Energy Load and Power Forecasting
Hosei University
2020-2024
Changzhou Institute of Technology
2016
Compressive Sensing (CS) surpasses the limitations of sampling theorem by reducing signal dimensions during sampling. Recent works integrate measurement coding into CS to enhance compression ratio. However, these significantly decrease image quality, and both encoding decoding become time-consuming. This paper proposes a based Image Codec with Partial Pre-calculation (CSCP) solve issues. The CSCP separates original reconstruction procedure two parts: reconstructing frequency domain data...
In compressed sensing (CS) based CMOS image sensors (CS-CIS), the ternary measurement matrix determines compression performance in terms of decoded quality versus sampling rate (data rate). Several studies have been carried out to investigate effect Hadamard and Walsh projection order selection on reconstruction by simply reordering orthogonal matrices [1]. However, there is still room for improvement reconstructed images from these works, especially at low SR. this paper, we propose a...
Dual learning trains two inverse processes tasks dually to further improve the selected tasks' performance. There are currently training paradigms in dual learning. One is directly utilize existing models for a manner models' However, it cannot effectively guarantee improvement of models. Another that networks both parties manually designed. Nevertheless, network performance will be poor initial stage training, which easily lead unsatisfactory training. Besides, most researches can only used...
The reconstruction of Compressed Sensing is iteration-based and contains numerous divisions, thereby costing tremendous processing time. In order to eliminate we adopt a sparse sensing matrix consisting mainly zero-vectors. After deleting these zero-vectors, an invertible full-rank obtained. Then the procedure can be replaced by multiplication operated in one iteration. Moreover, because inverse deterministic, simply processed shift add operators. proposed architecture verified on Xilinx...
Compressed sensing (CS), as a signal processing technique, is often used to acquire and reconstruct sparse signal. It can decrease the difficulty of acquiring while increase reconstructing Recently, block-based intra-prediction algorithms are widely further compression ratio images by using information neighboring blocks predict current block. However, it hard speed parallel due dependency among blocks. Meanwhile, reconstruction compressed time consuming. A algorithm Zigzag ordering-based...
Context Adaptive Binary Arithmetic Coding (CABAC) is the entropy coding module widely used in recent video standards such as HEVC/H.265 and VVC/H.266. CABAC a well-known throughput bottleneck due to its strong data dependencies. Because required context model of current bin often depends on results previous bin, cannot be prefetched early enough, then costs pipeline stalls. To solve this problem, we propose prediction-based prefetching strategy. If prediction correct, stalls can eliminated,...
Based on the concept of product image and relationship between brand image, through analysis five key elements in visual machine tool products, which are continuously applied to design process single series products family a system is consequently constructed end, practical significance enhancing enterprises’ market competitiveness.