An Improved Elbows Detection Algorithm for Underwater Blurred Images

Vehicle detection is an essential part of an intelligent traffic system, which is an important research field in drone application. Because unmanned aerial vehicles (UAVs) are rarely configured with stable camera platforms, aerial images are easily blurred. There is a challenge for detectors to accurately locate vehicles in blurred images in the target detection process. To improve the detection performance of blurred images, an end-to-end adaptive vehicle detection algorithm (DCNet) for drones is proposed in this article. First, the clarity evaluation module is used to determine adaptively whether the input image is a blurred image using improved information entropy. An improved GAN called Drone-GAN is proposed to enhance the vehicle features of blurred images. Extensive experiments were performed, the results of which show that the proposed method can detect both blurred and clear images well in poor environments (complex illumination and occlusion). The detector proposed achieves larger gains compared with SOTA detectors. The proposed method can enhance the vehicle feature details in blurred images effectively and improve the detection accuracy of blurred aerial images, which shows good performance with regard to resistance to shake.

Download Full-text

Research on Subway Pedestrian Detection Algorithm Based on Big Data Cleaning Technology

Wireless Communications and Mobile Computing ◽

10.1155/2021/4700204 ◽

2021 ◽

Vol 2021 ◽

pp. 1-10

Author(s):

Zhuoyang Lyu

Keyword(s):

Data Cleaning ◽

Pedestrian Detection ◽

Detection Algorithm ◽

Motion Blur ◽

Cleaning Process ◽

Detection Model ◽

Subway Stations ◽

Cleaning Technology ◽

Blurred Images

The pedestrian detection model has a high requirement on the quality of the dataset. Concerning this problem, this paper uses data cleaning technology to improve the quality of the dataset, so as to improve the performance of the pedestrian detection model. The dataset used in this paper is obtained from subway stations in Beijing and Nanjing. The data images’ quality is subject to motion blur, uneven illumination, and other noisy factors. Therefore, data cleaning is very important for this paper. The data cleaning process in this paper is divided into two parts: detection and correction. First, the whole dataset goes through blur detection, and the severely blurred images are filtered as the difficult samples. Then, the image is sent to DeblurGAN for deblur processing. 2D gamma function adaptive illumination correction algorithm is used to correct the subway pedestrian image. Then, the processed data is sent to the pedestrian detection model. Under different data cleaning datasets, through the analysis of the detection results, it is proved that the data cleaning process significantly improves the detection model’s performance.

Download Full-text

EDGE DETECTION IN ATOMIC MAGNETOMETER IMAGING

Journal of the Siena Academy of Sciences ◽

10.4081/jsas.2018.8532 ◽

2019 ◽

Vol 10 (1) ◽

Author(s):

N. Papi ◽

E. Cali ◽

C. Marinelli ◽

E. Mariotti ◽

V. Millucci

Keyword(s):

Edge Detection ◽

Electromagnetic Induction ◽

Room Temperature ◽

Signal To Noise Ratio ◽

Imaging Techniques ◽

Low Frequency ◽

Detection Algorithm ◽

X Rays ◽

Excellent Spatial Resolution ◽

Blurred Images

We present the results of an edge detection algorithm applied on Electromagnetic Induction Imaging provided by an Atomic radio-frequency Magnetometer operating in an unshielded environment and at room temperature. Atomic Magnetometers have been already used for Imaging Techniques in the last few years, but the image reconstruction and the object pattern recognition lacks nowadays in terms of quality: the effect of scattering of e.m. signals at low-frequency provides blurred images, and does not allow for a clean ray – optics response, as in the case of X rays. Our algorithm, based on solved Gaussian Noise Recognition, demonstrates excellent spatial resolution achieved despite low Signal-to-Noise-Ratio.

Download Full-text

Read, Understand, Learn, & Excel: Development and Testing of an Automated Reading Strategy Detection Algorithm for Postsecondary Students

American Journal of Speech-Language Pathology ◽

10.1044/2019_ajslp-18-0181 ◽

2019 ◽

Vol 28 (3) ◽

pp. 1257-1267 ◽

Cited By ~ 2

Author(s):

Priya Kucheria ◽

McKay Moore Sohlberg ◽

Jason Prideaux ◽

Stephen Fickas

Keyword(s):

Academic Success ◽

Undergraduate Students ◽

Expository Text ◽

Reading Strategy ◽

Detection Algorithm ◽

Expository Texts ◽

Human Observer ◽

Behavioral Strategies ◽

Periodic Review ◽

Wide Range

PurposeAn important predictor of postsecondary academic success is an individual's reading comprehension skills. Postsecondary readers apply a wide range of behavioral strategies to process text for learning purposes. Currently, no tools exist to detect a reader's use of strategies. The primary aim of this study was to develop Read, Understand, Learn, & Excel, an automated tool designed to detect reading strategy use and explore its accuracy in detecting strategies when students read digital, expository text.MethodAn iterative design was used to develop the computer algorithm for detecting 9 reading strategies. Twelve undergraduate students read 2 expository texts that were equated for length and complexity. A human observer documented the strategies employed by each reader, whereas the computer used digital sequences to detect the same strategies. Data were then coded and analyzed to determine agreement between the 2 sources of strategy detection (i.e., the computer and the observer).ResultsAgreement between the computer- and human-coded strategies was 75% or higher for 6 out of the 9 strategies. Only 3 out of the 9 strategies–previewing content, evaluating amount of remaining text, and periodic review and/or iterative summarizing–had less than 60% agreement.ConclusionRead, Understand, Learn, & Excel provides proof of concept that a reader's approach to engaging with academic text can be objectively and automatically captured. Clinical implications and suggestions to improve the sensitivity of the code are discussed.Supplemental Materialhttps://doi.org/10.23641/asha.8204786

Download Full-text