2023 International Conference on Images, Signals, and Computing

 

ICISC 2023

27-29 May 2023, Chengdu, China

Poster Papers

If you have questions for a poster, please contact the authors by email.

 

Paper Title

Highly reliable on-board Computer software design and verification for space radiation

Abstract

In this paper, the on-board computer software and its reliability design of a micro-satellite are analyzed. The micro-satellite on-board computer adopts the dual-computer redundancy method, and uses the dual-computer communication to realize the system operation state recovery and hold when the host and standby computer switch, and some reliability design of the computer hardware in the space environment is realized and completed by the software. The space flight test proves that the micro-satellite on-board computer software is safe and reliable. This paper introduces the main features of the micro-satellite’s On-board Data Handling software and the reliability design technology adopted, which reflects the micro-satellite’s On-Board Data Handling software’s main design technology in terms of reliability

Authors

  • ping wang ( sichuan vocational college of finance and economics ) < [email protected]>

 

Paper Title

Research on Methods of Entity Resolution in Dataspaces

Abstract

Dataspace is a new way of data integration. Entity resolution identifies two records that point to the same entity in the real world. In this paper, a record graph is constructed by using the records in the data set. The redundant comparisons are removed by pruning the record graph, and the records is divided into blocks according to the pruned graph. The subsequent entity resolution work is only carried out in blocks. When the entity is parsed in the block, the method of attribute mapping and expression representing attribute value is used to further divide the data to ensure the accuracy of parsing. Methods experiments were carried out on real data sets.

Authors

 

Paper Title

CYBER THREAT DETECTION, SECURING AND STORING CONFIDENTIAL FILES IN BYOD

Abstract

As the demand for Bring Your Own Device (BYOD) service enablement grows, it becomes an important element of the business. Because of the increased business agility, essential infrastructure is becoming more important. Employee productivity and happiness as a metric different way of working surroundings but harmful access to vital infrastructure from an unauthorized source. The use of BYOD has increased the number of cyber-attacks. As a result, it has become a major cybersecurity issue for most businesses. As a result, the corporate ecosystem fragments. There are numerous options. In recent years, new technologies and techniques for reducing pollution have been created. This section has been subjected to cyber attacks, which has resulted in the resolution of a number of concerns. However, due to new assault strategies and tools, work has resumed.is a condition that must be met on a regular basis. The majority of the company issuing a certificate-based authentication system that is secure to enable a bring-your-own-device (BYOD) environment However, the certificate-based system There are a slew of issues with the authentication system that need to be addressed. A security flaw that allows unauthorized access.

Authors

 

Paper Title

License plate recognition using Machine Learning

Abstract

Car owners altering license plates using different typefaces and designs violate the law that strictly forbids such behaviour. Traffic police officers claim that changing the license plates makes it impossible to read the registration numbers due to an increase in fatal street collisions and car thefts. They worry that it may be difficult to track down vehicles used in hit-and-run incidents. It is challenging to impose further limitations on any algorithm used to identify and recognise license plates in a developing nation like Bangladesh. This work has the primary objective of designing a reliable detection and recognition system for transitional, standard car license plates, which are frequently seen in developing countries. Increase the effectiveness of reading license plates drawn or printed in various styles and typefaces employing cutting-edge technology, including machine learning (ML) models. For this study, You Only Look Once (YOLOv3) is used to utilising the most recent version of the object detection method. The raw image is pre-processed to increase its quality and then divided into appropriate-sized grid cells to determine where the license plate should be placed. After that, the data is post-processed, and the accuracy of the proposed model is evaluted using industry-recognised standards. A sizeable image dataset was used to be tested using this proposed methodology. The presented system is expected to be essential for vehicle monitoring, parking fee collection, lowering traffic accidents, and identifying unregistered vehicles. The results demonstrate that the suggested method achieves 97.1% mAP, 95.3% precision and 96.8% in plate detection

Authors

 

Paper Title

Research on Multi-Target Vehicle Detection and Tracking Based on Yolo

Abstract

Vehicle detection and monitoring are gaining importance in traffic management. However, detection is still a n issue as vehicles vary in size, which directly affects vehicle counting accuracy. The proposed vehicle detection and counting method first extracts the road surface of the expressway in the image and divides it into distant regions. The newly developed segmentation strategy in the proposed vehicle identification and counting system first extracts The t rail’s state i n the picture furthermore separates it as far as the near areas. This method is important for improving vehicle detection. The above location is then sent to his YOLOv5m network to determine the vehicle type and location. Finally, we validate the proposed methodology using multiple traffic monitoring recordings f rom different environments. Also, the vehicle detection performance has increased to 99.39% of m ap compared to Yolov5 Basic. The research has practical consequences for the management and control of vehicle objects in traffic situations.

Authors

 

Paper Title

Research on visual navigation based on remote sensing image

Abstract

At present, the positioning and navigation system covers global positioning and navigation system (GPS) and inertial navigation system, but GPS navigation and inertial navigation have their own shortcomings. Visual navigation can perceive and provide rich dynamic environment information, and can realize accurate and reliable positioning and navigation in more complex environments. In this paper, a remote sensing image feature matching algorithm suitable for image navigation and positioning is constructed. Aiming at the problems of difficult and time-consuming matching of remote sensing images obtained by aircraft sensors, the key points of fast feature are extracted, the main and secondary features are selected according to the standard, the direction intensity histogram is calculated by using the relative distance, azimuth and relationship intensity, and the descriptor is constructed through the calculation results. Cosine similarity is used as the similarity measure of directional intensity histogram to match the features in the two images to achieve feature matching. In the filtering error matching stage, the error matching on the hypothesis set is removed by calculating the good matching with high embedding ratio, and the matching result with high accuracy is obtained. Based on this algorithm, a navigation prototype system based on
remote sensing image is designed and implemented to provide a visualization
platform.

Authors

  • nan Li ( School of Computer science, Northwestern Polytechnical University ) < [email protected]>

 

Paper Title

Diabetic Retinopathy Classification through fundus images using deep learning

Abstract

One of the most common eye diseases in the
people aged between 20-74 years is Diabetic Retinopathy (DR).
DR is an eye complication where the patient loses his vision due
to an increase in glucose levels in the blood. DR is most
prominent in the patients who are diagnosed with diabetic
mellitus. Over one-third of the diabetic mellitus patients are
diagnosed with DR. For diagnosing DR, the patient has to visit
an ophthalmologist for dilated eye examination. However,
everyone cannot have this facility. Hence, there is a need for a
simple automated software for diagnosing the five stages of DR
efficiently. In this paper, a simple model is developed using the
Kaggle APTOS Blindness Detection dataset which is publicly
available. In the pre-processing step the images are enhanced
and the deep learning model ResNet152 architecture is used for
the classification step. After training the ReseNet152 model
yielded a training and validation loss of 0.073 and 0.107
respectively and validation accuracy of 0.984. Further, a simple
Graphical User Interface is developed using tkinter framework
in python standard library which classifies the given input
fundus image as one of the five stages of DR.
Keywords—Diabetic Retinopathy, Deep Learning, ResNet152,
Training Loss, Validation Loss, Validation accuracy, Graphical
User Interface.

Authors

 

Paper Title

Low-Complexity Moving Object Detection Algorithm in Dynamic Background

Abstract

The scale of the monitoring system is becoming larger and larger. In order to perform intelligent video processing in surveillance systems, we need to detect moving object in image sequence. Some methods in the literature can achieve a valid detection result, but usually they have high computational complexity. In the outdoor scenes, the background is usually dynamic, and the dynamic background makes it difficult to detect moving object. In order to solve these problems, we propose a new method with low computational complexity using mass center coordinate to expand the mask image. The proposed method can remove the interference of dynamic background in the detection. Experiment results show that our method can mask the dynamic background more completely while ensuring fast computation and consuming less hardware resources. The method can be used in massive video intelligent processing.

Authors

 

Paper Title

Adolescent Dysmorphic Disorder Model Research based on Machine Learning

Abstract

Nowadays, dysmorphic disorder among contemporary adolescents has attracted more and more attention from people of all social circles. The purpose of this study is to provide a useful self-evaluation model of adolescent image for assessing adolescents’ dysmorphic disorder situations. 249 teenagers participated in this study and various machine learning algorithms have been developed and utilized for building the self-evaluation model, such as the K-Nearest Neighbor algorithm, Naïve Bayes algorithm, and Principal Component Analysis algorithm. The best self-evaluation model developed in this project gave the highest accuracy of 76.92% on the testing set. For predicting the trend of dysmorphic disorder among contemporary Chinese adolescents, ordinary least squares linear regression model has been created, and then the percentages of different age stages to carry out major plastic surgery in 2022, 2023, and 2024 have been predicted.

Authors

  • Leyao Bi ( The High School Affiliated to Renmin University of Chin ) < [email protected]>

 

Paper Title

global temporal pyramid for human abnormal action recognition

Abstract

With the development of monitoring technology and the improvement of people’s security awareness, intelligent human abnormal action recognition technology in the field of action recognition is increasingly high. In most cases, abnormal human action may have little difference in appearance compared with normal behavior, so the control of visual rhythm information becomes an important factor affecting action recognition, but people often focus on the appearance information of the action and ignore the rhythm information. In this paper, we introduce the temporal pyramid module to process the visual tempos information, meanwhile, the traditional LSTM local history information transfer method is very easy to lose the context information, which is not conducive to the grasp of global information and thus will greatly affect the processing effect of the temporal pyramid. This paper introduces a non-local neural network module to enhance the network’s ability to grasp global information and the model’s long-range modeling capability, which is used to supplement the temporal pyramid module. Finally, this paper uses the mainstream anomaly dataset UCF-Crime to test the network performance, and the improved network model recognition accuracy AUC reaches 0.82, which is better than other state-of-the-art methods.

Authors

 

Paper Title

Prediction of Cardiovascular disease based on Image Segmentation of CT Scans

Abstract

Machine learning is often considered an advanced technology that only highly trained experts can access. This prevents many doctors and biologists from using this tool in their research. This article aims to dispel that outdated notion. Heart failure (HF) has been shown to be one of the most common causes of death, so accurate and timely prediction of the risks of heart failure is necessary and of utmost importance. Cardiovascular diseases are a broad category of several diseases that affect the heart and blood vessels. Early predictive methods for cardiovascular disease helped make decisions about changes in at-risk patients that reduced risks. The healthcare industry contains a lot of medical data, so machine learning algorithms are needed to predict heart disease. We propose using U-net and deep learning to segment cardiac CT scan images
to check the area of the heart that may cause/ is causing problems in the foreseeable future/currently. A comparative analysis was made of selected publications that were published in the years 2016 to 2022.

Authors

 

Paper Title

Large-field gesture tracking and recognition for augmented reality interaction

Abstract

In recent years, with the continuous development of computer vision and artificial intelligence technology, gesture recognition is widely used in many fields, such as virtual reality, augmented reality and so on. However, the traditional binocular camera architecture is limited by its limited field of view Angle and depth perception range. Fisheye camera is gradually applied in gesture recognition field because of its advantage of larger field of view Angle. Fisheye cameras offer a wider field of vision than previous binocular cameras, allowing for a greater range of gesture recognition. This gives fisheye cameras a distinct advantage in situations that require a wide field of view. However, because the imaging mode of fisheye camera is different from traditional camera, the image of fisheye camera has a certain degree of distortion, which makes the calculation of gesture recognition more complicated.Our goal is to design a distortion correction processing strategy suitable for fisheye cameras in order to extend the range of gesture recognition and achieve large field of view gesture recognition. Combined with binocular technology, we can use the acquired hand depth information to enrich the means of interaction. By taking advantage of the large viewing Angle of the fisheye camera to expand the range of gesture recognition, make it more extensive and accurate. This will help improve the real-time and precision of gesture recognition, which has important implications for artificial intelligence, virtual reality and augmented reality.

Authors

 

Paper Title

The evaluation and optimization of port efficiency

Abstract

This paper mainly aims at the evaluation and optimization of port efficiency, reasonably sets the key performance indicators of personnel, technology and process, and builds the AHP model, the fuzzy comprehensive evaluation model, the cost efficiency balance optimization model and the effectiveness evaluation model of port efficiency based on the ideas of AHP, fuzzy mathematics and mathematical programming. To solve the problem of evaluation and optimization. Building the model and implementing the model with concrete and visible data helps us solve our problems.

Authors

 

Paper Title

Development of thickness measurement software for aircraft absorbing coatings based on magnetic force measurement

Abstract

In this paper, a magnetic-force based thickness measurement model was established based on the physical properties of radar absorbing coatings that are easy to be magnetized and the experimental data obtained from the standard samples with titanium alloy substrate coatings. The experimental data obtained from the thickness measurement experiment of the standard samples with titanium alloy substrate coatings were fitted by using least squares method. A thickness measurement sensor was developed based on the STM32 microcontroller, which solved the problem that the traditional wireless design was unable to balance transmission distance, anti-interference, and power consumption. On this basis, thickness measurement software was developed based on WinCE 6.0 and measurement experiments were done. The experimental results showed that this magnetic-force based thick measurement system had the advantages of simple structure and high software reliability.

Authors

 

Paper Title

GPU-based parallel optimization strategy for Barrett’s algorithm

Abstract

With the development of computer science, hardware devices have emerged, which greatly enhance the efficiency of computing and data processing. GPU, as a highly parallelized computing chip, can be used in areas such as personal needs and scientific computing.Additionally, advances in cryptography enable data to have more effective security; and there are some computational bottlenecks in cryptographic algorithms during practical use. The use of GPUs to accelerate cryptographic algorithms is one of the hot topics of research in the industry and academia today.
RSA algorithm is the most widely used asymmetric cryptography public-private key system, which guarantees the security of the public-private key generated by the algorithm through the difficulty of large number decomposition. In the actual encryption and decryption process of RSA, substantial large number power modulo operations are involved, which seriously affects the computing efficiency of RSA and a set of optimization strategies are urgently needed to shorten the computing time of RSA.
In RSA algorithm, there is parallel optimization space for modulo operation. In this paper, the structural optimization of the algorithm is realized based on Montgomery algorithm, and the algorithm’s arithmetic optimization is realized by using Barrett reduction algorithm to achieve the algorithm-level tuning. And the optimization algorithm is deployed on the GPU side through CUDA architecture.
By comparing with the large number library GMP running on the CPU, the algorithm optimization and parallelization deployment in this paper achieves significant speedup in the RSA encryption and decryption operations. In this paper, the performance analysis reveals that there is room for further optimization in terms of program structure and memory usage.

Authors

 

Paper Title

Implementation of Edge Detection

Abstract

Edge is a set of connected pixels which lie on the boundary between two regions which differ in grey value. The pixels on the edge are called the edge points. Edge detection plays a very important role in the analysis of different characteristics and properties of the image. When an edge is detected with the help of an Edge detection system, the unnecessary details are removed, while only the important structural information is retained. There are various Edge Detection Algorithms and Techniques available such as Robert, Prewitt, Sobel, Marr-Hildrith, Canny etc. which are used in the processing of images. This paper reviews all these gradient techniques and gives a brief analysis of all the Techniques. MATLAB is used as the software tool for the purpose of conversion of a RGB image to Grayscale and then produces a text file which contains all the pixel values of the image. The Xilinx ISE is used for the purpose of synthesis and configuring. Verilog HDL is used to design the whole system. The Hardware Description Language Code is implemented and synthesized on the Spartan 3E FPGA board.

Authors

 

Paper Title

Merging Public Opinion Information and Stock Numerical Data for Stock Trend Prediction Based on Deep Learning

Abstract

Unlike other stock markets participants, the participants in China mainland are composed of individual investors, which account for 82% of the trading volume of the stock market. The decision-making basis of individual investors is mainly public opinion and recent stock prices. Therefore, the public opinion on professional stock social sites has an important impact on the decision of individual investors, which in turn affects the trend of the stock market. However, the previous stock market forecasting methods mostly ignored the influence of public opinion information on the market. For this reason, this paper proposes a novel framework to predict the stock trend by using both public opinion and stock numerical data. The original contributions of this paper include stock commentary word embedding model based on the stock comment text data crawled from https://xueqiu.com through two-stage training and LSTM-CNN layered model based on the improved self-attention mechanism. Two main experiments are conducted: the first experiment extract stock commentary word embedding, and the second experiment forecasts the stock price trends of Shanghai and Shenzhen A-share market. Results show that: 1)LSTM-CNN layered model is better than previous methods; 2)The combination of public opinion information and numerical data can improve the performance of the model; 3)Stock commentary word embedding model is better than pre-training word embedding model; 4) The longer the data span, the better the stock forecasting model will perform.

Authors

 

Paper Title

Improved k-means-based FAKM Clustering Method for Scientific and Technical Literature

Abstract

Research on rapid clustering technology based on bibliographic information of scientific and technical literature aims to efficiently realize the correlation analysis of scientific and technical literature, laying the foundation for discovering hot spots and trends in the research field, conducting interdisciplinary and cross-border research, and accurately recommending scientific and technical literature. Focusing on the analysis of clustering algorithms, we proposed an improved k-means-based Firefly Algorithm k-means (FAKM) clustering method, which effectively solved the problem of randomly selecting the initial center points of class cluster when using k-means algorithm for clustering in the clustering stage, which leads to local optimum, low accuracy and large gap between the division of class clusters and the real situation of clustering results. The use of FAKM clustering algorithm resulted in better clustering performance, high accuracy, and fewer iterations. The experimental results showed that the method achieved a silhouette coefficient of 0.54 and adjust mutual information of 0.69 on the same scientific and technical literature data set, which proved the good performance of the method.

Authors

 

Paper Title

Lightweight Facial Emotion Recognition Network Based on Inception V3

Abstract

[Objective]Facial emotion recognition is mainly used in psychological medicine, driving monitoring, human-computer interaction and so an. In recent years, Convolutional Neural Network is a common method for facial emotion recognition. With the extensive and in-depth application, the performance requirement of facial emotion recognition is increasing day by day. Augmenting the depth and width of the network is more direct method to improve the recognition performance, but it will cause the parameters and computation to increase rapidly and lead to a long training time. In order to solve this problem, a lightweight face emotion recognition network based on Inception V3 is proposed. The Shortcoming of long training time due to too many parameters in original Inception V3 is improved. [Method(detailed)] The network consists of two convolutional layers, two asymmetric Inception modules, two pooling layers and two fully connected layers combined with Dropout. The total parameters of proposed model is 70% of Inception V3. The FFA-NET filter image is used to enhance the image and the Gaussian noise is used to balance the dataset. The accuracy of the model is improved by dynamic attenuation learning rate of the loss value in validation set. [Result] RAF-DB, CK+ and JAFFE data sets are used to train the model. The training time of a single image is less than 16ms. The average recognition accuracy on testing set is 77.89% and 91.82% respectively. [Conclusion]Compared with the algorithms with superior performance in face emotion recognition in recent years, the accuracy has been improved and the training time is shorter, which can be better applied to various scenarios.

Authors

 

Paper Title

Reducing Ringing Artefact in Fresnel Digital Holography Using Compressed Sensing

Abstract

Compressed sensing is a signal processing technique used for signal reconstruction with significantly smaller number of samples than the requirements of the Nyquist-Shannon theorem. In this work, we simulate a lenseless digital holographic system. We investigate the ringing-like artefact introduced by truncation by the camera aperture. We present the results of using the orthogonal matching pursuit based compressed sensing algorithms to combat this ringing-like artefact. We demonstrate that compressed sensing achieves remarkable reconstructions and suppresses ringing well, but only up to a point in terms of the size of the aperture. This research could help the advancement of compressive digital holography.

Authors

 

Paper Title

Wireless and Sensors Network Security Threads and Count Measures

Abstract

Wireless sensor network is domain-specific wireless system composed of various nodes called sensor nodes. The purpose of the nodes is to collect physical and environmental data parameters such as sound, pressure and temperature, and relay the data across the network through a central node. The goal is to get the data from the sensors in a secure manner. Data flow in wireless sensor networks is threatened by a variety of attack types, including Sybil, Wormhole, Sinkhole, and others. In order to accomplish privacy, integrity, availability, confidentiality number of protocols and techniques have been developed. In this paper we will provide a systematic review that spot the light on the common threats and the countermeasures proposed by the previous studies.

Authors

 

Paper Title

Research on a Personalized Classifier of Health Status Based on Pulse Signal

Abstract

At present, the workload of mental workers in society is getting heavier and heavier, and it is necessary to assess their health status. Compared with other physiological signals, the pulse is easy to obtain and non-invasive. In this paper, through pulse signal detection, pulse data preprocessing and feature extraction, 12 sets of feature values are selected. Then based on these feature data, using support vector machine algorithm modeling, for different testers to build different personalized human physiological state discrimination system. The experimental results show that the classification accuracy rate reaches 91.17%, which proves that the selected feature value has a strong correlation with the physiological state, and the classifier is effective.

Authors

  • Xiaoping Jiang ( China University of Mining and Technology (Beijing) School of Mechanical and Electrical engineering ) < [email protected]>
  • Leli Sun ( China University of Mining and Technology (Beijing) School of Mechanical and Electrical engineering ) < [email protected]>
  • Shuyao Feng ( China University of Mining and Technology (Beijing) School of Mechanical and Electrical engineering ) < [email protected]>
  • Zhuojing Li ( China University of Mining and Technology (Beijing) School of Mechanical and Electrical engineering ) < [email protected]>
  • Ying Chen ( China University of Mining and Technology (Beijing) School of Mechanical and Electrical engineering ) < [email protected]>
  • Xingzhuo Chen ( China University of Mining and Technology (Beijing) School of Mechanical and Electrical engineering ) < [email protected]>
  • Chunman Wang ( China University of Mining and Technology (Beijing) School of Mechanical and Electrical engineering ) < [email protected]>
  • Aolai He ( China University of Mining and Technology (Beijing) School of Mechanical and Electrical engineering ) < [email protected]>

 

Paper Title

A high-resolution image dehazing GAN model in icing meteorological environment

Abstract

In this paper, we propose a high-resolution GAN model for image dehazing in icing meteorological environment, which strictly follows a physics-driven scattering strategy. First of all, the utilization of sub-pixel convolution realizes the model to remove image artifacts and generate high-resolution images. Secondly, we use Patch-GAN in the discriminator to drive the generator to generate a haze-free image by capturing the details and local information of the image. Furthermore, to restore the texture information of the foggy image and reduce color distortion, the model is jointly trained by multiple loss functions. Experiments show the proposed method achieves advanced performance for image dehazing in icing weather environment.

Authors

Scroll to Top