Efficient In-Place Hough Transform Algorithm for Arbitrary Image Sizes

We consider the problem of scheduling jobs with equal lengths and arbitrary sizes on uniform parallel batch machines with different capacities. Each machine can only process the jobs whose sizes are not larger than its capacity. Several jobs can be processed as a batch simultaneously on a machine, as long as their total size does not exceed the machine’s capacity. The objective is to minimize makespan. Under a divisibility constraint, we obtain two efficient exact algorithms. For the general problem, we obtain an efficient 2-approximation algorithm. Previous work has shown that the problem cannot be approximated to within an approximation ratio better than 2, unless P = NP, even when all machines have identical speeds and capacities.

Signal flow graph approach to efficient and forward stable DST algorithms

The via assignment problem, whose objective is to minimize the number of via columns used, is known to be NP-hard even when each net includes at most three vias. The authors develop efficient approximation algorithms for solving this problem. An algorithm for the case in which the size of each net is bounded by three is presented, and is then extended to the case in which the nets have arbitrary sizes. >

Convolutional neural networks (CNNs), the most important deep learning networks for computer vision, have undergone a series of developments and improvements for image-related tasks such as object recognition, image classification, semantic segmentation, etc. However, in the field of natural language processing (NLP), the novel attention-based network Transformer had a profound impact on machine translation, which subsequently led to a boom in attention-based models for computer vision. State-of-the-art models with attention have already shown good performance for computer vision tasks due to the sophisticated design of network architectures and advanced computational efficiency techniques. For example, selfattention learns relationships between segments or words in different positions compared to the current performance of convolutional neural networks. Inspired by Vision Transformer (ViT), we propose a simple novel transformer architecture model, called Flexible Transformer, which inherits the properties of attention-based architectures and is flexible for inputs of arbitrary size. Besides self-attention, the inputs in ViT are not pre-processed, such as resizing or cropping, but kept intact without altering them, which could lead to distortion or loss of information. In this paper, we want to present a novel and simple architecture that meets these requirements. Compared to state-of-arts, our model processes inputs with arbitrary image sizes without any pre-processing and pretraining costs. Also, the results of the experiments show that the model can potentially provide good results with high accuracy despite limited resources. Even though the results of Flexible Transformer are not as accurate as those of Vision Transformer, they show the potential of a model with high performance in image classification tasks with variable size images. The significance of the research opens up possibilities for dealing with primitive images in deep learning tasks. Based on the original inputs, reliable results with good accuracy could be obtained if the proposed model is optimized and further trained on large datasets.

As a cutting-edge research direction in the field of radar imaging, video SAR has the capability of high-resolution and persistent imaging at any time and under any weather. Video SAR requires high computational efficiency of the imaging algorithm, and PFA has become the preferred imaging algorithm because of its applicability to the spotlight mode and relatively high computational efficiency. However, traditional PFA also has problems, such as low efficiency and limited scene size. To address the above problems, a generalized persistent polar format algorithm, called GPPFA, is proposed for airborne video SAR imaging that is applicable to the persistent imaging requirements of airborne video SAR under multitasking conditions. Firstly, the wavenumber domain resampling characteristics of video SAR PFA are analyzed, and a generalized resampling method is proposed to obtain higher efficiency. Secondly, for the problem of scene size limitation caused by wavefront curvature error, an efficient compensation method applicable to different scene sizes is proposed. GPPFA is capable of video SAR imaging at different wavebands, different slant ranges, and arbitrary scene sizes. Point target and extended target experiments verify the effectiveness and efficiency of the proposed method.

Many species have an essentially continuous distribution in space, in which there are no natural divisions between randomly mating subpopulations. Yet, the standard approach to modelling these populations is to impose an arbitrary grid of demes, adjusting deme sizes and migration rates in an attempt to capture the important features of the population. Such indirect methods are required because of the failure of the classical models of isolation by distance, which have been shown to have major technical flaws. A recently introduced model of extinction and recolonisation in two dimensions solves these technical problems, and provides a rigorous technical foundation for the study of populations evolving in a spatial continuum. The coalescent process for this model is simply stated, but direct simulation is very inefficient for large neighbourhood sizes. We present efficient and exact algorithms to simulate this coalescent process for arbitrary sample sizes and numbers of loci, and analyse these algorithms in detail.

Real-time non-line-of-sight imaging is crucial for practical applications. Among existing methods, transient methods present the best visual reconstruction ability. However, most transient methods require a long acquisition time, thus failing to deal with real-time imaging tasks. Here, we provide a dual optical coupling model to describe the spatiotemporal propagation of photons in free space, then propose an efficient non-confocal transformation algorithm and establish a non-confocal time-to-space boundary migration model. Based on these, a scan-free boundary migration method is proposed. The data acquisition speed of the method can reach 151 fps, which is ∼7 times faster than the current fastest data acquisition method, while the overall imaging speed can also reach 19 fps. The background stability brought by fast scan-free acquisition makes the method suitable for dynamic scenes. In addition, the high robustness of the model to noise makes the method have the capability of non-line-of-sight imaging in outdoor environments during the daytime. To further enhance the practicality of this method in real-world scenarios, we exploit the statistical prior and propose a plug-in-and-play super-resolution method to extract higher spatial resolution signals, reducing the detector array requirement from 32 × 32 to 8 × 8 without compromising imaging quality, thus reducing the device expense of detectors.

This paper presents a novel locality sensitive histogram (LSH) algorithm for visual tracking. Unlike the conventional image histogram that counts the frequency of occurrence of each intensity value by adding ones to the corresponding bin, an LSH is computed at each pixel location, and a floating-point value is added to the corresponding bin for each occurrence of an intensity value. The floating-point value exponentially reduces with respect to the distance to the pixel location where the histogram is computed. An efficient algorithm is proposed that enables the LSHs to be computed in time linear in the image size and the number of bins. In addition, this efficient algorithm can be extended to exploit color images. A robust tracking framework based on the LSHs is proposed, which consists of two main components: a new feature for tracking that is robust to illumination change and a novel multiregion tracking algorithm that runs in real time even with hundreds of regions. Extensive experiments demonstrate that the proposed tracking framework outperforms the state-of-the-art methods in challenging scenarios, especially when the illumination changes dramatically. Evaluation using the latest benchmark shows that our algorithm is the top performer.

Over the last four years, Essex Corporation has been developing an image reconstruction system (ImSyn<SUP>TM</SUP>) for a variety of applications, including synthetic aperture radar, magnetic resonance imaging, and other sensors. Each of these formats encompasses its own set of requirements for image quality, image size, and throughput. The ImSyn system has been designed to provide image dynamic range of up to 100 dB, arbitrary image sizes, and throughputs of 30 images per second to satisfy the various requirements. Recent efforts to enhance the ImSyn breadboard version has required the development of modeling techniques to measure system performance and identify sources of error and artifacts. The modeling approach, application of the measurements to determine and improve system performance, and the resulting imagery achieved with the ImSyn breadboard are presented.

Community search has aroused widespread interest in the past decades. Among existing solutions, the learning-based models exhibit outstanding performance in terms of accuracy by leveraging labels to 1) train the model for community score learning, and 2) select the optimal threshold for community identification. However, labeled data are not always available in real-world scenarios. To address this notable limitation of learning-based models, we propose a pre-trained graph Trans former based community search framework that uses Zero label (i.e., unsupervised), termed TransZero. TransZero has two key phases, i.e., the offline pre-training phase and the online search phase. Specifically, in the offline pre-training phase, we design an efficient and effective community search graph transformer ( CSGphormer ) to learn node representation. To pre-train CSGphormer without the usage of labels, we introduce two self-supervised losses, i.e., personalization loss and link loss, motivated by the inherent uniqueness of node and graph topology, respectively. In the online search phase, with the representation learned by the pre-trained CSGphormer , we compute the community score without using labels by measuring the similarity of representations between the query nodes and the nodes in the graph. To free the framework from the usage of a label-based threshold, we define a new function named expected score gain to guide the community identification process. Furthermore, we propose two efficient and effective algorithms for the community identification process that run without the usage of labels. Extensive experiments over 10 public datasets illustrate the superior performance of TransZero regarding both accuracy and efficiency.

While exhibiting superior performance in many tasks, vision transformers (ViTs) face challenges in quantization. Some existing low-bit-width quantization techniques cannot effectively cover the whole inference process of ViTs, leading to an additional memory overhead (22.3%-172.6%) compared with corresponding fully quantized models. To address this issue, we propose quadruplet uniform quantization (QUQ) to deal with data of various distributions in ViT. QUQ divides the entire data range into at most four subranges that are uniformly quantized with different scale factors. To determine the partition scheme and quantization parameters, an efficient relaxation algorithm is proposed accordingly. Moreover, dedicated encoding and decoding strategies are devised to facilitate the design of an efficient accelerator. Experimental results show that QUQ surpasses state-of-the-art quantization techniques; it is the first viable scheme that can fully quantize ViTs to 6-bit with acceptable accuracy. Compared with conventional uniform quantization, QUQ leads to not only a higher accuracy but also an accelerator with lower area and power.

Efficient regeneration and transformation procedures have been developed with mature tissues from apricot cultivars. Integration of transgenes (gfp and nptII marker genes) in the plant cells have been demonstrated with transformed callus tissues by means of observation of GFP expression, exposure to high kanamycin doses, PCR and Southern blot hybridization. However, to date, production of transformed plants had not been reported. This work was aimed to develop a strategy based on antibiotic selection that allowed the production of the first transformed apricot plants from mature leaf tissues. Application of two different aminoglycoside antibiotics at different concentrations and at different times during the transformation experiments have been tried. High selection pressure with 40 μM kanamycin failed to produced transformed buds, so antibiotic concentration was reduced to 25.7 μM and applied right after co-culture, delayed for 5, 7 or 10 days or not applied. However, only one transformed bud was obtained when kanamycin was applied immediately after co-culture and two chimeras when no selection was applied. Given the results, different aminoglycoside antibiotics were tested at different concentrations and 40 μM paromo-mycine was found to better stimulate growth of transformed tissues. Therefore, this antibiotic was used in subsequent experiments, although when applied at that concentration from the beginning of the experiments, transformed plants were not obtained. A progressive selection strategy has been necessary to obtain transformed apricot plants. When a low selection pressure (10 μM) was applied during the first 10 days after co-culture some transformed buds were obtained but selection efficiency was really low and many escapes were also produced. Increasing the initial paromomycin concentration to 20 μM followed by a higher selection applied after 10 to 24 days after co-culture have greatly improved transformation and selection efficiencies. Based on GFP expression and PCR, some of these strategies have produced over 7% of transformed buds.

A method for producing large numbers of transgenic wheat plants has been developed. With this approach, an average of 9.7% of immature embryo explants were transformed and generated multiple self-fertile, independently transformed plants. No untransformed plants, or escapes, were regenerated. This transformation procedure uses morphogenic calli derived from scutellum tissue of immature embryos of Triticum aestivum cv. Bobwhite co-bombarded with separate plasmids carrying a selectable marker gene (bar) and a gene of interest, respectively. Transformed wheat calli with a vigorous growth phenotype were obtained by extended culture on media containing 5.0 mg/l bialaphos. These calli retained morphogenic potential and were competent for plant regeneration for as long as 11 months. The bar gene and the gene of interest were co-expressed in T0 progeny plants. This wheat transformation protocol may facilitate quantitative production of multiple transgenic plants and significantly reduce the cost and labor otherwise required for screening out untransformed escapes.

We describe a highly efficient and reproducible Agrobacterium-mediated transformation protocol applicable to several varieties of tomato (Solanum lycopersicum, earlier known as Lycopersicum esculentum). Conditions such as co-cultivation period, bacterial concentration, concentration of benzyl amino purine (BAP), zeatin and indole acetic acid (IAA) were optimized. Co-cultivation of explants with a bacterial concentration of 108 cells/ml for three days on 2 mg/l BAP, followed by regeneration on a medium containing 1 mg/ml zeatin resulted in a transformation frequency of 41.4%. Transformation of tomato plants was confirmed by Southern blot analysis and beta-glucuronidase (GUS) assay. The protocol developed showed very high efficiency of transformation for tomato varieties Pusa Ruby, Arka Vikas and Sioux. The optimized transformation procedure is simple, efficient and does not require tobacco, Petunia, tomato suspension feeder layer or acetosyringone.

Penicillium marneffei is an opportunistic fungal pathogen of humans and the only dimorphic species identified in its genus. At 25 degrees P. marneffei exhibits true filamentous growth, while at 37 degrees P. marneffei undergoes a dimorphic transition to produce uninucleate yeast cells that divide by fission. Members of the STE12 family of regulators are involved in controlling mating and yeast-hyphal transitions in a number of fungi. We have cloned a homolog of the S. cerevisiae STE12 gene from P. marneffei, stlA, which is highly conserved. The stlA gene, along with the A. nidulans steA and Cryptococcus neoformans STE12alpha genes, form a distinct subclass of STE12 homologs that have a C2H2 zinc-finger motif in addition to the homeobox domain that defines STE12 genes. To examine the function of stlA in P. marneffei, we isolated a number of mutants in the P. marneffei-type strain and, in combination with selectable markers, developed a highly efficient DNA-mediated transformation procedure and gene deletion strategy. Deletion of the stlA gene had no detectable effect on vegetative growth, asexual development, or dimorphic switching in P. marneffei. Despite the lack of a detectable function, the P. marneffei stlA gene complemented the sexual defect of an A. nidulans steA mutant. In addition, substitution rate estimates indicate that there is a significant bias against nonsynonymous substitutions. These data suggest that P. marneffei may have a previously unidentified cryptic sexual cycle.