Code Regions Research Articles

An Adaptive Deep Learning Framework for Shipping Container Code Localization and Recognition

Shipping containers play an important role in global transportation. As container codes are the unique identifiers for shipping containers, recognizing these codes is an essential step to manage the containers and logistics. The conventional code localization methods can easily be interfered by varied noises and cannot identify the best regions for code recognition. In this article, we propose an adaptive deep learning framework for shipping container code localization and recognition. In the framework, the noisy text regions will be removed by an adaptive score aggregation (ASA) algorithm. The code region boundaries are identified by the average-to-maximum suppression range (AMSR) algorithm. Thus, the predicted locations can be adjusted within this range to fit the code recognition model to achieve higher accuracy. The experimental results on the comparative study with the state-of-the-art models, including EAST, PSENet, GCRNN, and MaskTextSpotter, demonstrated that the proposed framework achieved better localization performance and obtained 93.33% recognition accuracy. The processing speed reaches 1.13 frames/s, which is sufficient to meet the operational requirements. Thus, the proposed solution will facilitate the digital transformation of shipping container management and logistics at ports.

Zeroploit

In this article, we first characterize register operand value locality in shader programs of modern gaming applications and observe that there is a high likelihood of one of the register operands of several multiply, logical-and, and similar operations being zero, dynamically. We provide intuition, examples, and a quantitative characterization for how zeros originate dynamically in these programs. Next, we show that this dynamic behavior can be gainfully exploited with a profile-guided code optimization called Zeroploit that transforms targeted code regions into a zero-(value-)specialized fast path and a default slow path. The fast path benefits from zero-specialization in two ways, namely: (a) the backward slice of the other operand of a given multiply or logical-and can be skipped dynamically, provided the only use of that other operand is in the given instruction, and (b) the forward slice of instructions originating at the given instruction can be zero-specialized, potentially triggering further backward slice specializations from operations of that forward slice as well. Such specialization helps the fast path avoid redundant dynamic computations as well as memory fetches, while the fast-slow versioning transform helps preserve functional correctness. With an offline value profiler and manually optimized shader programs, we demonstrate that Zeroploit is able to achieve an average speedup of 35.8% for targeted shader programs, amounting to an average frame-rate speedup of 2.8% across a collection of modern gaming applications on an NVIDIA® GeForce RTX™ 2080 GPU.

XMeter: Finding Approximable Functions and Predicting Their Accuracy

Approximate computing has significant potential to improve the efficiency of a computing system. Numerous techniques have been proposed in literature. Virtually, all of them require programmers to either experiment with every instance of a specific type of code region exhaustively to find approximable code regions or annotate such regions manually. Both approaches are error-prone and can lead to missed opportunities. Therefore, we propose XMeter to automatically find and quantify approximable code regions. XMeter, first, analyzes the application code statically using a novel algorithm based on memory location updates. Also, XMeter provides a deep learning-based predictor to predict the accuracy of the application when different code regions are approximated. Our proposed scheme does not require the programmer to experiment exhaustively for all possible error rates and types of approximation techniques. Moreover, the scheme does not require any domain knowledge and is not specific to any approximation technique. Therefore, it is general enough to be applicable for any approximation technique. We developed XMeter using LLVM and experimented with 10 applications. We analyzed 43 approximable functions and found 21 to be highly tolerant of errors. We validated our results using 4 well-known approximation techniques and showed that XMeter can predict an application's accuracy accurately.

Fast identification method for express end sorting label code based on convolutional recurrent neural network

In the automatic sorting process of express, the express end sorting label code is used to indicate that the express is dispatched to a specific address by a specific courier. Since there are many areas on the express bill containing digital information, some areas may be improperly photographed, etc. The difficulty in positioning and recognizing the express end sorting label code region is increased. To solve this problem, this paper proposes an express end sorting label code recognition method with convolutional recurrent neural network for the code specification, which has certain versatility. In order to improve the overall code recognition speed, this paper optimizes the traditional digital recognition method, removes the original segmentation operation of the character and recognizes the code as sequence recognition. Firstly, the coding region is located, and then, the express end sorting label code is recognized by the convolutional recurrent neural network. In order to test the experimental performance, this paper tests on Free-Type dataset and SUN-synthesized dataset. The experimental results show that the proposed method improves the recognition accuracy and processing speed of the express end sorting label code.

Database of Romanian cave invertebrates with a Red List of cave species and a list of hotspot/coldspot caves.

BackgroundThe increasing human impact in Romanian caves raises the urgency of publishing a correct database of the strictly-adapted cave fauna. Previous attempts at indexing cave fauna and classifying caves by using their fauna opened many questions regarding the use of an incomplete list of cave species and mixed lists of troglobionts/stygobionts with troglophiles/stygophiles for ranking caves with priority for protection. It has also become obvious that there is a need to publish a list of Romanian cave species that are under threat. Cave species in Romania (and elsewhere) are endemic on small ranges, are unique and must be considered as important units for conservation. A cave must be equally protected if it has one or more rare and strictly endemic cave species. Although not exhaustive, we here provide the first checklist of Romanian troglobionts/stygobionts developed in the framework of the DARKFOOD and GROUNDWATERISK projects, coordinated by the “Emil Racovita” Institute of Speleology, Cluj-Napoca, Romania. The GIS application was used to complement the checklist of cave species with data on caves and surface environments above the caves. Until complete data on species diversity and population sizes are made available for each cave, measures of conservation can be implemented, based on the presence/absence of cave species, while classifications of caves for protection, based on the number of species, must be avoided. We also propose a list of Romanian caves with fauna that are under threat and a tentative Red List of Romanian troglobiont/stygobionts.New informationThis is the first database with identified troglobiont and stygobiont species of Romania, with a critical analysis of their distribution inside the country. A list of caves that need protection for their rare and unique species and a tentative Red List of Romanian cave fauna are also added. A total of 173 species were identified, of which 77 troglobionts and 96 stygobionts are currently registered in 366 caves. The database is divided into two parts, one part with a list of troglobionts, their revised systematic position, cave name, cave code and geographic region; and the second part with the same information on stygobionts. The database represents the contribution of many active researchers, who are the authors of this paper and of review publications of many other authors of the "Emil Racoviță" Institute of Speleology.

Psc2code

Programming screencasts have become a pervasive resource on the Internet, which help developers learn new programming technologies or skills. The source code in programming screencasts is an important and valuable information for developers. But the streaming nature of programming screencasts (i.e., a sequence of screen-captured images) limits the ways that developers can interact with the source code in the screencasts. Many studies use the Optical Character Recognition (OCR) technique to convert screen images (also referred to as video frames) into textual content, which can then be indexed and searched easily. However, noisy screen images significantly affect the quality of source code extracted by OCR, for example, no-code frames (e.g., PowerPoint slides, web pages of API specification), non-code regions (e.g., Package Explorer view, Console view), and noisy code regions with code in completion suggestion popups. Furthermore, due to the code characteristics (e.g., long compound identifiers like ItemListener), even professional OCR tools cannot extract source code without errors from screen images. The noisy OCRed source code will negatively affect the downstream applications, such as the effective search and navigation of the source code content in programming screencasts. In this article, we propose an approach named psc2code to denoise the process of extracting source code from programming screencasts. First, psc2code leverages the Convolutional Neural Network (CNN) based image classification to remove non-code and noisy-code frames. Then, psc2code performs edge detection and clustering-based image segmentation to detect sub-windows in a code frame, and based on the detected sub-windows, it identifies and crops the screen region that is most likely to be a code editor. Finally, psc2code calls the API of a professional OCR tool to extract source code from the cropped code regions and leverages the OCRed cross-frame information in the programming screencast and the statistical language model of a large corpus of source code to correct errors in the OCRed source code. We conduct an experiment on 1,142 programming screencasts from YouTube. We find that our CNN-based image classification technique can effectively remove the non-code and noisy-code frames, which achieves an F1-score of 0.95 on the valid code frames. We also find that psc2code can significantly improve the quality of the OCRed source code by truly correcting about half of incorrectly OCRed words. Based on the source code denoised by psc2code , we implement two applications: (1) a programming screencast search engine; (2) an interaction-enhanced programming screencast watching tool. Based on the source code extracted from the 1,142 collected programming screencasts, our experiments show that our programming screencast search engine achieves the precision@5, 10, and 20 of 0.93, 0.81, and 0.63, respectively. We also conduct a user study of our interaction-enhanced programming screencast watching tool with 10 participants. This user study shows that our interaction-enhanced watching tool can help participants learn the knowledge in the programming video more efficiently and effectively.

Socioeconomic disparities in SUDEP in the US.

To determine the impact of socioeconomic status (SES) on sudden unexpected death in epilepsy (SUDEP) rates. We queried all decedents presented for medico-legal investigation at 3 medical examiner (ME) offices across the country (New York City, Maryland, San Diego County) in 2009 to 2010 and 2014 to 2015. We identified all decedents for whom epilepsy/seizure was listed as cause/contributor to death or comorbid condition on the death certificate. We then reviewed all available reports. Decedents determined to have SUDEP were included for analysis. We used median income in the ZIP code of residence as a surrogate for SES. For each region, zip code regions were ranked by median household income and divided into quartiles based on total population for 2 time periods. Region-, age-, and income-adjusted epilepsy prevalence was estimated in each zip code. SUDEP rates in the highest and lowest SES quartiles were evaluated to determine disparity. Examined SUDEP rates in 2 time periods were also compared. There were 159 and 43 SUDEP cases in the lowest and highest SES quartiles. ME-investigated SUDEP rate ratio between the lowest and highest SES quartiles was 2.6 (95% confidence interval [CI] 1.7-4.1, p < 0.0001) in 2009 to 2010 and 3.3 (95% CI 1.9-6.0, p < 0.0001) in 2014 to 2015. There was a significant decline in overall SUDEP rate between the 2 study periods (36% decrease, 95% CI 22%-48%, p < 0.0001). ME-investigated SUDEP incidence was significantly higher in people with the lowest SES compared to the highest SES. The difference persisted over a 5-year period despite decreased overall SUDEP rates.

Regional variation of care dependency after hip fracture in Germany: A retrospective cohort study using health insurance claims data.

ObjectiveTo investigate variation of care dependency after hip fracture across German regions based on the assessment by the German statutory long-term care insurance.Data sources/study settingPatient-level statutory health and long-term care insurance claims data from 2009–2011 and official statistical data from Germany.Study designWe performed a retrospective cohort study. Investigated multinomial outcome categories were increase in care dependency (new onset or a higher care dependency than pre-fracture), no change as reference and death as competing risk in the quarterly period following hip fracture (follow-up 3 months). Regional variation was operationalized with the variance of regional-level random intercepts based on generalized linear mixed models. We adjusted for patient and regional characteristics.Principal findingsThe study included 122,887 hip fracture patients in 95 German postal code regions. Crude outcomes were 30.87% increase in care dependency and 14.35% death. Results indicated modest variation on regional level. Male sex, increasing age, increasing comorbidity, pertrochanteric and subtrochanteric fracture site compared to femoral neck, time from hospital admission to surgery of 3 or more days, as well as increasing inpatient length of stay, non-participation in rehabilitation and regions with lower hospital density were positively associated with an increase in care dependency.ConclusionsSeveral characteristics on patient and regional level associated with the outcome were identified. Variation in the increase in care dependency after hip fracture appeared to be attributable primarily to patient characteristics. Variation on regional level was only modest.

Analysis of Sinus Balloon Catheter Dilation Providers Based on Medicare Provider Utilization and Payment Data.

The use of balloon catheter dilation (BCD) to treat chronic rhinosinusitis has increased dramatically since its conception, necessitating further characterization of BCD providers and trends in its usage. Medicare data on BCD providers have made it possible to study recent demographic patterns. There has also been an increase in mid-level providers' scope of otolaryngologic practice that is not well defined. To better understand BCD adoption by studying volume of BCD procedures as well as training, geography, and practice socioeconomic characteristics of BCD providers for Medicare beneficiaries. We reviewed Medicare Provider Utilization and Payment Data Public Use Files for 2014 and 2015 for providers with claims for BCD of the sinuses. We extracted provider zip code, state, gender, and number of services per BCD code. We obtained median household income by zip code and geographic region based on US Census Bureau data. Providers were classified using an Internet search to determine practice setting and type of specialty training/certification. In 2014 and 2015, 428 providers performed 42 494 BCDs billed to Medicare beneficiaries. Among BCD providers, 5.1% were female, 98.1% had Doctor of Medicine/Doctor of Osteopathic Medicine credentials, and 1.9% had nurse practitioner/physician assistant credentials. Over the 2-year period, the median number of BCDs was 63 for physicians and 37 for mid-level providers. Fellowship-trained rhinologists performed a median of 38 BCDs over 2 years. The most common subspecialty certification/training was in facial plastics and reconstructive surgery. The majority of providers (63.8%) performed 1 to 99 BCDs over the 2 years. In the South, there were 21.9 BCD procedures performed per 100 000 people compared to 7.3 in the Northeast, 9.3 in the Midwest, and 8.5 in the West. There is a large range in total BCD procedures performed by individual providers, and this varies by certain provider characteristics. Mid-level providers have emerged as a significant population performing BCD.

Repeatable determinism using non-random weight initialisations in smart city applications of deep learning

Modern Smart City applications draw on the need for requirements that are safe, reliable and sustainable, as such these applications have a need to utilise machine-learning mechanisms such that they are consistent with public liability. Machine and deep learning networks, therefore, are required to be in a form that is safe and deterministic in their development and also in their deployment. The viability of non-random weight initialisation schemes in neural networks make the network more deterministic in learning sessions which is a desirable property in safety critical systems where deep learning is applied to smart city applications and where public liability is a concern. The paper uses a variety of schemes over number ranges and gradients and achieved a 98.09% accuracy figure, + 0.126% higher than the original random number scheme at 97.964%. The paper highlights that in this case, it is the number range and not the gradient that is affecting the achieved accuracy most dominantly, although there can be a coupling of number range with activation functions used. Unexpectedly in this paper, an effect of numerical instability was discovered from run to run when run on a multi-core CPU. The paper also has shown the enforcement of consistent deterministic results on an multi-core CPU by defining atomic critical code regions, and that aids repeatable information assurance in model fitting (or learning sessions). That enforcement of consistent repeatable determinism has also a benefit to accuracy even for the random schemes, and a highest score of 98.29%, + 0.326% higher than the baseline was achieved. However, also the non-random initialisation scheme causes weight arrangements after learning to be more structured which has benefits for validation in safety critical applications.

Implementing Risk Management in Pervasive and IoT Environments

The pervasive nature of networked things envision various risks as these digital devices generate high volume of data with variable nature. The technological growth is also a product of highly interrelated complex data, so it becomes a strong argument for risks management in pervasive and internet of things environments. This research analyzes many risks present in pervasive and IoT environments. The paper elaborates various risk analysis strategies in the pervasive and IoT environments which are highly configurable in nature. The paper has implemented risk management in pervasive applications by providing risky code insights by a smart software. The risky regions of software code are analyzed by the software and managed on priority. The state of art constructs a strong case for establishing interrelationships between risks management and quality assurance in big computation environments.

I/O dependent idempotence bugs in intermittent systems

Intermittently-powered, energy-harvesting devices operate on energy collected from their environment and must operate intermittently as energy is available. Runtime systems for such devices often rely on checkpoints or redo-logs to save execution state between power cycles, causing arbitrary code regions to re-execute on reboot. Any non-idempotent program behavior—behavior that can change on each execution—can lead to incorrect results. This work investigates non-idempotent behavior caused by repeating I/O operations, not addressed by prior work. If such operations affect a control statement or address of a memory update, they can cause programs to take different paths or write to different memory locations on re-executions, resulting in inconsistent memory states. We provide the first characterization of input-dependent idempotence bugs and develop IBIS-S, a program analysis tool for detecting such bugs at compile time, and IBIS-D, a dynamic information flow tracker to detect bugs at runtime. These tools use taint propagation to determine the reach of input. IBIS-S searches for code patterns leading to inconsistent memory updates, while IBIS-D detects concrete memory inconsistencies. We evaluate IBIS on embedded system drivers and applications. IBIS can detect I/O-dependent idempotence bugs, giving few (IBIS-S) or no (IBIS-D) false positives and providing actionable bug reports. These bugs are common in sensor-driven applications and are not fixed by existing intermittent systems.

Distance transform and template matching based methods for localization of barcodes and QR codes

Visual codes play an important role in automatic identification, which became an inseparable part of industrial processes. Thanks to the revolution of smartphones and telecommunication, it also becomes more and more popular in everyday life, containing embedded web addresses or other small informative texts. While barcode reading is straightforward in images having optimal parameters (focus, illumination, code orientation, and position), localization of code regions is still challenging in many scenarios. Every setup has its own characteristics, therefore many approaches are justifiable. Industrial applications are likely to have more fixed parameters like illumination, camera type and code size, and processing speed and accuracy are the most important requirements. In everyday use, like with smartphone cameras, a wide variety of code types, sizes, noise levels and blurring can be observed, but the processing speed is often not crucial, and the image acquisition process can be repeated in order for successful detection. In this paper, we address this problem with two novel methods for localization of 1D barcodes based on template matching and distance transformation, and a third method to detect QR codes. Our proposed approaches can simultaneously localize several different types of codes. We compare the effectiveness of the proposed methods with several approaches from the literature using public databases and a large set of synthetic images as a benchmark. The evaluation shows that the proposed methods are efficient, having 84.3% Jaccard accuracy, superior to other approaches. One of the presented approaches is an improvement on our previous work. Our template matching based method is computationally more complex, however, it can be adapted to specific code types providing high accuracy. The other method uses distance transformation, which is fast and gives rough regions of interests that can contain valid visual code candidates.

Global budget versus cost ceiling: a natural experiment in hospital payment reform in the Netherlands

Global budget (GB) arrangements have become a popular method worldwide to control the rise in healthcare expenditures. By guaranteeing hospital funding, payers hope to eliminate the drive for increased production, and incentivize providers to deliver more efficient care and lower utilization. We evaluated the introduction of GB contracts by certain large insurers in Dutch hospital care in 2012 and compared health care utilization to those insurers who continued with more traditional production-based contracts, i.e., cost ceiling (CC) contracts. We used the share of GB hospital funding per postal code region to study the effect of contract types. Our findings show that having higher share of GB financing was associated with lower growth in treatment intensity, but it was also associated with higher growth in the probability of having at least one hospital visit. While the former finding is in line with our expectation, the latter is not and suggests that hospital visits may take longer to respond to contract incentives. Our study covers the years of 2010–2013 (2 years before and 2 years following the introduction of the new contracts). Therefore, our results capture only short-term effects.

Towards Gradually Typed Capabilities in the Pi-Calculus

Gradual typing is an approach to integrating static and dynamic typing within the same language, and puts the programmer in control of which regions of code are type checked at compile-time and which are type checked at run-time. In this paper, we focus on the pi-calculus equipped with types for the modeling of input-output capabilities of channels. We present our preliminary work towards a gradually typed version of this calculus. We present a type system, a cast insertion procedure that automatically inserts run-time checks, and an operational semantics of a pi-calculus that handles casts on channels. Although we do not claim any theoretical results on our formulations, we demonstrate our calculus with an example and discuss our future plans.

Managing code transformations for better performance portability

Code optimization is an intricate task that is getting more complex as computing systems evolve. Managing the program optimization process, including the implementation and evaluation of code variants, is tedious, inefficient, and errors are likely to be introduced in the process. Moreover, because each platform typically requires a different sequence of transformations to fully harness its computing power, the optimization process complexity grows as new platforms are adopted. To address these issues, systems and frameworks have been proposed to automate the code optimization process. They, however, have not been widely adopted and are primarily used by experts with deep knowledge about underlying architecture and compiler intricacies. This article describes the requirements that we believe necessary for making automatic performance tuning more broadly used, especially in complex, long-lived high-performance computing applications. Besides discussing limitations of current systems and strategies to overcome these, we describe the design of a system that is able to semi-automatically generate efficient platform-specific code. In the proposed system, the code optimization is programmer-guided, separately from application code, on an external file in what we call optimization programming. The language to program the optimization process is able to represent complex collections of transformations and, as a result, generate efficient platform-specific code. A database manages different optimized versions of code regions, providing a pragmatic approach to performance portability, and the framework itself has separate components, allowing the optimized code to be used on systems without installing all of the modules required for the code generation. We present experiments on two different platforms to illustrate the generation of efficient platform-specific code that performs comparable to hand-optimized, vendor-provided code.

Do Employees From Less-Healthy Communities Use More Care and Cost More? Seeking to Establish a Business Case for Investment in Community Health.

IntroductionFew studies have examined the impact of community health on employers. We explored whether employed adults and their adult dependents living in less-healthy communities in the greater Philadelphia region used more care and incurred higher costs to employers than employees from healthier communities. MethodsWe used a multi-employer database to identify adult employees and dependents with continuous employment and mapped them to 31 zip code regions. We calculated community health scores at the regional level, by using metrics similar to the Robert Wood Johnson Foundation (RWJF) County Health Rankings but with local data. We used descriptive analyses and multilevel linear modeling to explore relationships between community health and 3 outcome variables: emergency department (ED) use, hospital use, and paid claims. Business leaders reviewed findings and offered insights on preparedness to invest in community health improvement.ResultsPoorer community health was associated with high use of ED services, after controlling for age and sex. After including a summary measure of racial composition at the zip code region level, the relationship between community health and ED use became nonsignificant. No significant relationships between community health and hospitalizations or paid claims were identified. Business leaders expressed interest in further understanding health needs of communities where their employees live.ConclusionThe health of communities in which adult employees and dependents live was associated with ED use, but similar relationships were not seen for hospitalizations or paid claims. This finding suggests a need for more primary care access. Despite limited quantitative evidence, business leaders expressed interest in guidance on investing in community health improvement.

A journey among Java neutral program variants

Neutral program variants are alternative implementations of a program, yet equivalent with respect to the test suite. Techniques such as approximate computing or genetic improvement share the intuition that potential for enhancements lies in these acceptable behavioral differences (e.g., enhanced performance or reliability). Yet, the automatic synthesis of neutral program variants, through program transformations remains a key challenge. This work aims at characterizing plastic code regions in Java programs, i.e., the code regions that are modifiable while maintaining functional correctness, according to a test suite. Our empirical study relies on automatic variations of 6 real-world Java programs. First, we transform these programs with three state-of-the-art program transformations: add, replace and delete statements. We get a pool of 23,445 neutral variants, from which we gather the following novel insights: developers naturally write code that supports fine-grain behavioral changes; statement deletion is a surprisingly effective program transformation; high-level design decisions, such as the choice of a data structure, are natural points that can evolve while keeping functionality. Second, we design 3 novel program transformations, targeted at specific plastic regions. New experiments reveal that respectively 60%, 58% and 73% of the synthesized variants (175,688 in total) are neutral and exhibit execution traces that are different from the original.

Efficient QR code authentication mechanism based on Sudoku

QR code is an important means for delivering information which has been widely used in our daily life. As an ISO international standard, the QR code encoding and decoding process are disclosed publicly, thus it is easy to decode a QR code then forge a new QR code with the same QR code public message. It can lead to the problems of information forgery and ease the spreading of fake news. To overcome this weakness, we propose a simple and efficient QR code authentication mechanism to embed the authentication information in the padding region of QR code based on the characteristics of Sudoku and Reed-Solomon code. Different from the previous scheme, the proposed scheme embeds the authentication information without consuming the QR code error correction capacity and is able to achieve a higher embedding capacity. Experimental results show that the proposed scheme has high security, low power consumption and is robust to common QR code attacks.

Formally verified software countermeasures for control-flow integrity of smart card C code

Fault attacks can target smart card programs to disrupt an execution and take control of the data or the embedded functionalities. Among all possible attacks, control-flow attacks aim at disrupting the normal execution flow. Identifying harmful control-flow attacks and designing countermeasures at the software level are tedious and tricky for developers. In this paper, we propose a methodology to detect harmful inter- and intra-procedural jump attacks at the source code level and automatically inject formally proven countermeasures into a C code. The proposed software countermeasures protect the integrity of individual statements at the granularity of single C statements. They support many control-flow constructs of the C language. The countermeasure scheme can detect an attack early either inside a control-flow construct or only at its exit. The secured source code defeats 100% of attacks that jump over at least two C source code statements. Experiments showed that the resulting code is also hardened against unexpected function calls and jump attacks at the assembly code level. Securing a source code automatically and extensively with our scheme degrades the performance. The performance overhead of our countermeasures on three well-known encryption algorithms available in C ranged from +41% to +138% on an x86 platform and from +45% to +217% on an ARM-v7 platform. However, combining code rewriting with hardening of sensitive code regions identified by the weakness detection step enables an application to be fully hardened while limiting the overhead.