Recent advances in computer vision---in the form of deep neural networks---have made it possible to query increasing volumes of video data with high accuracy. However, neural network inference is computationally expensive at scale: applying a state-of-the-art object detector in real time (i.e., 30+ frames per second) to a single video requires a $4000 GPU. In response, we present NoScope, a system for querying videos that can reduce the cost of neural network video analysis by up to three orders of magnitude via inference-optimized model search. Given a target video, object to detect, and reference neural network, NoScope automatically searches for and trains a sequence, or cascade, of models that preserves the accuracy of the reference network but is specialized to the target video and are therefore far less computationally expensive. NoScope cascades two types of models: specialized models that forego the full generality of the reference model but faithfully mimic its behavior for the target video and object; and difference detectors that highlight temporal differences across frames. We show that the optimal cascade architecture differs across videos and objects, so NoScope uses an efficient cost-based optimizer to search across models and cascades. With this approach, NoScope achieves two to three order of magnitude speed-ups (265-15,500x real-time) on binary classification tasks over fixed-angle webcam and surveillance video while maintaining accuracy within 1--5% of state-of-the-art neural networks.

NoScope: optimizing neural network queries over video at scale

Despite the large number of both commercial and academic methods for Automatic License Plate Recognition (ALPR), most existing approaches are focused on a specific license plate (LP) region (e.g. European, US, Brazilian, Taiwanese, etc.), and frequently explore datasets containing approximately frontal images. This work proposes a complete ALPR system focusing on unconstrained capture scenarios, where the LP might be considerably distorted due to oblique views. Our main contribution is the introduction of a novel Convolutional Neural Network (CNN) capable of detecting and rectifying multiple distorted license plates in a single image, which are fed to an Optical Character Recognition (OCR) method to obtain the final result. As an additional contribution, we also present manual annotations for a challenging set of LP images from different regions and acquisition conditions. Our experimental results indicate that the proposed method, without any parameter adaptation or fine tuning for a specific scenario, performs similarly to state-of-the-art commercial systems in traditional scenarios, and outperforms both academic and commercial approaches in challenging ones.

/pdf/license-plate-detection-and-recognition-in-unconstrained-1xyjsrjklj.pdf

License Plate Detection and Recognition in Unconstrained Scenarios

In this paper, we tackle the problem of car license plate detection and recognition in natural scene images. We propose a unified deep neural network, which can localize license plates and recognize the letters simultaneously in a single forward pass. The whole network can be trained end-to-end. In contrast to existing approaches which take license plate detection and recognition as two separate tasks and settle them step by step, our method jointly solves these two tasks by a single network. It not only avoids intermediate error accumulation but also accelerates the processing speed. For performance evaluation, four data sets including images captured from various scenes under different conditions are tested. Extensive experiments show the effectiveness and the efficiency of our proposed approach.

Toward End-to-End Car License Plate Detection and Recognition With Deep Neural Networks

This paper presents a robust and efficient method for license plate detection with the purpose of accurately localizing vehicle license plates from complex scenes in real time. A simple yet effective image downscaling method is first proposed to substantially accelerate license plate localization without sacrificing detection performance compared with that achieved using the original image. Furthermore, a novel line density filter approach is proposed to extract candidate regions, thereby significantly reducing the area to be analyzed for license plate localization. Moreover, a cascaded license plate classifier based on linear support vector machines using color saliency features is introduced to identify the true license plate from among the candidate regions. For performance evaluation, a data set consisting of 3977 images captured from diverse scenes under different conditions is also presented. Extensive experiments on the widely used Caltech license plate data set and our newly introduced data set demonstrate that the proposed approach substantially outperforms state-of-the-art methods in terms of both detection accuracy and run-time efficiency, increasing the detection ratio from 91.09% to 96.62% while decreasing the run time from 672 to 42 ms for processing an image with a resolution of $1082\times 728$ . The executable code and our collected data set are publicly available.

A Robust and Efficient Approach to License Plate Detection

Nowadays, the railway industry is in a position where it is able to exploit the opportunities created by the IIoT (Industrial Internet of Things) and enabling communication technologies under the paradigm of Internet of Trains. This review details the evolution of communication technologies since the deployment of GSM-R, describing the main alternatives and how railway requirements, specifications and recommendations have evolved over time. The advantages of the latest generation of broadband communication systems (e.g., LTE, 5G, IEEE 802.11ad) and the emergence of Wireless Sensor Networks (WSNs) for the railway environment are also explained together with the strategic roadmap to ensure a smooth migration from GSM-R. Furthermore, this survey focuses on providing a holistic approach, identifying scenarios and architectures where railways could leverage better commercial IIoT capabilities. After reviewing the main industrial developments, short and medium-term IIoT-enabled services for smart railways are evaluated. Then, it is analyzed the latest research on predictive maintenance, smart infrastructure, advanced monitoring of assets, video surveillance systems, railway operations, Passenger and Freight Information Systems (PIS/FIS), train control systems, safety assurance, signaling systems, cyber security and energy efficiency. Overall, it can be stated that the aim of this article is to provide a detailed examination of the state-of-the-art of different technologies and services that will revolutionize the railway industry and will allow for confronting today challenges.

https://www.mdpi.com/1424-8220/17/6/1457/pdf

Towards the Internet of Smart Trains: A Review on Industrial IoT-Connected Railways

Video-based traffic flow monitoring is a fast emerging field based on the continuous development of computer vision. A survey of the state-of-the-art video processing techniques in traffic flow monitoring is presented in this paper. Firstly, vehicle detection is the first step of video processing and detection methods are classified into background modeling based methods and non-background modeling based methods. In particular, nighttime detection is more challenging due to bad illumination and sensitivity to light. Then tracking techniques, including 3D model-based, region-based, active contour-based and feature-based tracking, are presented. A variety of algorithms including MeanShift algorithm, Kalman Filter and Particle Filter are applied in tracking process. In addition, shadow detection and vehicles occlusion bring much trouble into vehicle detection, tracking and so on. Based on the aforementioned video processing techniques, discussion on behavior understanding including traffic incident detection is carried out. Finally, key challenges in traffic flow monitoring are discussed.

Video processing techniques for traffic flow monitoring: A survey

Traffic surveillance is an important topic in intelligent transportation systems. Robust vehicle detection and tracking is one challenging problem for complex urban traffic surveillance. This paper proposes a rear-view vehicle detection and tracking method based on multiple vehicle salient parts using a stationary camera. We show that spatial modeling of these vehicle parts is crucial for overall performance. First, the vehicle is treated as an object composed of multiple salient parts, including the license plate and rear lamps. These parts are localized using their distinctive color, texture, and region feature. Furthermore, the detected parts are treated as graph nodes to construct a probabilistic graph using a Markov random field model. After that, the marginal posterior of each part is inferred using loopy belief propagation to get final vehicle detection. Finally, the vehicles' trajectories are estimated using a Kalman filter, and a tracking-based detection technique is realized. Experiments in practical urban scenarios are carried out under various weather conditions. It can be shown that our method adapts to partial occlusion and various lighting conditions. Experiments also show that our method can achieve real-time performance.

Rear-View Vehicle Detection and Tracking by Combining Multiple Parts for Complex Urban Surveillance

This paper presents a novel algorithm for license plate detection in complex scenes, particularly for the all-day traffic surveillance environment. Unlike low-level feature-based methods, our work is motivated by component-based models for object detection. The detection process is divided into three steps, namely, decomposition, modeling, and inference. First, observing that one license plate is decomposed into several constituent characters, the maximally stable extremal region detector is used to extract candidate characters in images. Then, conditional random field (CRF) models are constructed on the candidate characters in neighborhoods. This way, the spatial and visual relationships among the characters is integrated in CRF in the form of probability distribution. Finally, the exact bounding boxes of license plates are estimated through the belief propagation inference on CRF. Both visual and structural features of license plates are fully exploited during detection. Hence, our approach can adapt to various environmental factors, such as cluttered background and illumination variation. A series of experiments are conducted on images that are collected from the actual road surveillance environment. The experimental results show the outstanding detection performance of the proposed method comparing with traditional algorithms.

Component-Based License Plate Detection Using Conditional Random Field Model

Due to FPGA's flexibility and parallelism, it is popular for accelerating image processing. In this paper, a double-parallel architecture based on FPGA has been exploited to speed up median filter and edge detection tasks, which are essential steps during image processing. The double-parallel scheme includes an image-level parallel and an operation-level parallel. The image-level parallel is a high-level parallel which divides one image into different parts and processes them concurrently. The operation-level parallel, which is embedded in each image-level parallel thread, fully explores every parallel part inside the concrete algorithms. The corresponding design is based on a DE2 Development Board which contains a CYCLONE II FPGA device. Meanwhile, the same task has also been implemented on PC and DSP for performance comparison. Despite the fact that operating frequencies of used PC and DSP are much higher than FPGA's, FPGA costs less time per computed image than both of them. By taking advantage of the double-parallel technique, the speed/frequency ratio of FPGA is 202 times faster than PC and 147 times faster than DSP. Finally, a detailed discussion about different platforms is conducted, which analyzes advantages and disadvantages of used computing platforms. This paper reveals that the proposed double-parallel scheme can dramatically speed up image processing methods even on a low-cost FPGA platform with low frequency and limited resources, which is very meaningful for practical applications.

Fast double-parallel image processing based on FPGA

In complex urban traffic conditions, the accurate detection of vehicles is challenging to current vehicle detection methods. To achieve the precise vehicle detection in complex urban traffic conditions, we have proposed a vehicle detection method based on a deformable hybrid image template in this paper. Our method contains two steps: constructing our hybrid image template and its probability model, and detecting vehicles from traffic images by using a three-stage SUM-MAX procedure. In the template construction step, small image patches constituting the hybrid image template are automatically learned from training images. These image patches have various features and the feature types include sketch, texture, flatness, and color. Furthermore, a probability model is proposed to assist the vehicle detection step. After the template construction, a three-stage SUM-MAX procedure is applied to achieve vehicle detection with local deformations in locations and orientations. There are two innovations in our method which are the applications of the hybrid image template and the three-stage SUM-MAX procedure in vehicle detection under complex urban traffic conditions. To evaluate our method, we have done a quantitative and contrast experiment and the experiment on complex urban traffic conditions. The experimental results show that our method can effectively cope with various vehicle poses, vehicle shapes, time-of-day and weather conditions. In particular, our method has good performance in complex urban traffic conditions.

Vehicle detection based on the deformable hybrid image template

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Ye Li

Author Tools

Chat about Author