Jingping Ma, Hui Xie, Longqing Li et al. · 2024
Automatic Emergency Braking(AEB) is a crucial method for enhancing vehicle driving safety. Uncertainty in road conditions, vehicle load, and braking system performance introduces disturbances to the AEB system's behavior. To address this issue, a self-optimizing cascade AEB control algorithm is proposed, which leverages multiple Extended State Observers (ESOs) to effectively mitigate these disturbances. First, upper and lower layer controllers are designed to plan and track the target deceleration respectively. Second, an long short-term memory-based three-dimensional classifier that considers three-axis attitude angle ESO data was designed to identify the vehicle's operating conditions, and the results are used in the cascade controller. Finally, through different condition simulation tests, the stopping distance error of the algorithm is less than 3.5cm. Compared to the standard algorithm, the algorithm effectively enhancing system stability with a precision improvement of 95.1%.
Cited by 0
Jaswinder Kaur, Sudhir Kumar Sharma · IITM Journal of Management and IT · 2018
Cascade-2 algorithm is a variant of well-known cascade-correlation algorithm that is widely investigated constructive training algorithm for designing cascade feed forward neural networks. This paper proposes a modified Cascade-2 algorithm with adaptive slope sigmoid function (MC2AASF). The algorithm emphasizes on architectural adaptation and functional adaptation during learning. This algorithm is a constructive approach of designing cascade architecture. To achieve functional adaptation, the slope of the sigmoid function is adapted during training. One simple variant is derived from MC2AASF is where the slope parameter of sigmoid function used at the hidden layers’ nodes is fixed to unity. Both the variants are compared to each other on five function approximation tasks. Simulation results show that adaptive slope sigmoid function presents several advantages over standard fixed shape sigmoid function, resulting in increasing flexibility, smoother learning, better generalization performance and better convergence
Cited by 0Open access
2000
Abstract. This paper considers some aspects of using a cascade-correlation network in the investment task in which it is required to determine the most suitable project to invest money. This task is one of the most often met economical tasks. In various bibliographical sources on economics there are described different methods of choosing investment projects. However, they all use either one or a few criteria, i.e., out of the set of criteria there are chosen most valuable ones. With this, a lot of information contained in other choice criteria is omitted. A neural network enables one to avoid information losses. It accumulates information and helps to gain better results when choosing an investment project in comparison with classical methods. The cascade-correlation network architecture that is used in this paper has been developed by Scott E. Fahlman and Cristian Lebiere at Carnegie Mellon University. Key words: neural network and criterion.
Cited by 0
Ola Bratteli, Palle E. T. Jørgensen · ArXiv.org · 1999
The spectral properties of the Ruelle transfer operator which arises from a given polynomial wavelet filter are related to the convergence question for the cascade algorithm for approximation of the corresponding wavelet scaling function.
Cited by 0Open access
FU Zheng-fang · 2010
Users can have access to resources and obtain services at anytime and anywhere in ubiquitous computing environment. By virtue of behavior experience in human society, according to dynamic evolution discipline of the contradiction between the lag of trust and the anteriority of service, and the self-organization, distributed computing and regenerative feedback characteristics of ant colony algorithm, cascade service selection model based on ant colony algorithm is put forward. It examines ubiquitous equipment' trust degree before ants choose the next hop, the best server satisfied with the ubiquitous entities' request was selected out by means of cascade servers to improve the service of quality and efficiency, which conquers the disadvantage that a singular equipment can't satisfy the request property and blindfold choice of service. Experimental results show ubiquitous entities can correctly select optimum network services in short time via this model.
Cited by 0
B. Harshini · 2024
Human Face Detection (FD)is a demanding topic in areas of image processing and pattern identification. Facial expression conveys non-verbal cues, which plays crucial role in social relations. Facial Expression is a significant yet challenging task, as we can use it to recognize emotions and mental state of individual. In this paper, using image processing and Machine Learning (ML), we differentiate the captured image with the trained dataset and then display the state of image. These handles detection and recognizing multiple faces in real time. Face Recognition (FR) can be done using two methods like FD and FR. This paper describes use of ML algorithms in real time which is used for FD and FR that is Haar Cascade algorithm.
Cited by 1
Ola Bratteli, Palle E. T. Jørgensen · Contemporary mathematics - American Mathematical Society · 1999
The spectral properties of the Ruelle transfer operator which arises from a given polynomial wavelet filter are related to the convergence question for the cascade algorithm for approximation of the corresponding wavelet scaling function.
Cited by 13
Heping Liu, Liu Yu, Lizhong Peng et al. · Progress in Natural Science Materials International · 2005
ABSTRACT A suitable correlation can be made to represent the simulated distillation of heavy oils starting from thennoaravimetric measurements. This method is applicable to hydrocarbons having an initial boiling point equal or greater than 200°C. The simulated thermogravimetric distillation fit was obtained from experiments with the standard compounds obtained from ASTM D-2887-83 (Boiling range distribution of petroleum fractions by gas chromatography). This method is simple, fast and without problems when applied to heavy feedstocks The data were used in the determination of average boiling temperatures of products from thermal cracking and thermal hydrocracking. It was also possible to quantify coke yields Average relative molecular masses of products from the above processes correlated well with the average boiling point temperatures. It indicates that, with respect to the hydrocarbon types, thermal cracking is not selective in comparison with thermal hydrocracking The equation applied to find the average boiling temperature is following: T = (1/al) (Ttg-a2). T is the boiling temperature, al and a2 are the correction factors, Ttg is the thermogravimetric temperature.
Cited by 7
Junfei Qiao, Fanjun Li, Hong-Gui Han et al. · Neurocomputing · 2015
Cited by 51
Dariusz Sychel, Aneta Bera, Przemysław Klęsk · Lecture notes in computer science · 2024
Cited by 0
Cheng, Yi, Liu, XiaoYan et al. · arXiv (Cornell University) · 2022
Cascade SVM (CSVM) can group datasets and train subsets in parallel, which greatly reduces the training time and memory consumption. However, the model accuracy obtained by using this method has some errors compared with direct training. In order to reduce the error, we analyze the causes of error in grouping training, and summarize the grouping without error under ideal conditions. A Balanced Cascade SVM (BCSVM) algorithm is proposed, which balances the sample proportion in the subset after grouping to ensure that the sample proportion in the subset is the same as the original dataset. At the same time, it proves that the accuracy of the model obtained by BCSVM algorithm is higher than that of CSVM. Finally, two common datasets are used for experimental verification, and the results show that the accuracy error obtained by using BCSVM algorithm is reduced from 1% of CSVM to 0.1%, which is reduced by an order of magnitude.
Cited by 0Open access
Juan Pablo Bello, Marcel L. Hernandez, Robert C. Simon · 2024
This paper presents the development and evaluation of a web application for facial detection and image processing, designed to profile senior citizens in Naga City, Philippines. The system integrates the Haar Cascade Algorithm and OpenCV, advanced facial recognition technologies, to automate data entry, reduce human error, and enhance data security in barangay services. The study follows the Iterative Model in the System Development Life Cycle (SDLC), refining the system's functionality through multiple iterations. A usability test conducted at Barangay Calauag assessed the system's performance in terms of usability, accuracy, reliability, and security. Initial results revealed significant challenges in system reliability, particularly due to uncontrolled lighting conditions. However, following system enhancements, reliability improved, and strong agreement was observed across all evaluation metrics. The final system demonstrated a notable increase in profiling efficiency, a significant reduction in human error, and improved security measures for senior citizen transactions. These findings confirm that the system can streamline data collection, enhance service delivery, and improve overall user satisfaction. The paper suggests continued collaboration with local authorities to optimize system performance, refine facial recognition under varying conditions, and ensure long-term support for the senior population of Naga City.
Cited by 0
A. H. Land, S. W. Stairs · Management Science · 1967
Large graphs for which the calculation of the shortest distances between all vertices is required often consist of several parts with limited interconnections. This paper shows how this characteristic may be used to reduce computer storage for a matrix type calculation, to reduce calculations] labour, and to provide results in a form which is particularly convenient if a subsequent assignment of commodity flow to shortest paths is needed. The technique presented here allows the partition to be arbitrary, save for a single, easily achieved, condition.
Cited by 39
Hitanshu A. Patel, Ritesh D. Rajput · 2018
In today's world, everything is transforming into automation. From basic households work to major industries work, from basic machinery to steering and stabilization of ships, aircraft and other applications and vehicles. Our main objective of this project is to control surveillance system completely without human intervention not only in steady or particular allotted or settled area but which can move to remote places and restricted area where human entry is not feasible. We have demonstrated our idea using a moving car with a mounted camera and processor and which can move autonomously or can be controlled remotely via remote/website/android app. The images which are captured by a camera can be processed by a processor and digital image processing (DIP) algorithms can be done for human detection or object detection. Boosted cascade, Haar cascade, anisotropic diffusion, Hausdorff distance, colour processing, background subtraction statistics of parts, pulse coupled neural networks, single correlation filter, these are various algorithms used for human/object detection.
Cited by 7
Wei Liu, Linjin Wu, Yuan Hu et al. · Marine Geodesy · 2024
Cited by 2
Dandan Qiu, Jianguo Ju, Shumin Ren et al. · Frontiers in Oncology · 2024
Pancreatic tumors are small in size, diverse in shape, and have low contrast and high texture similarity with surrounding tissue. As a result, the segmentation model is easily confused by complex and changeable background information, leading to inaccurate positioning of small targets and false positives and false negatives. Therefore, we design a cascaded pancreatic tumor segmentation algorithm. In the first stage, we use a general multi-scale U-Net to segment the pancreas, and we exploit a multi-scale segmentation network based on non-local localization and focusing modules to segment pancreatic tumors in the second stage. The non-local localization module learns channel and spatial position information, searches for the approximate area where the pancreatic tumor is located from a global perspective, and obtains the initial segmentation results. The focusing module conducts context exploration based on foreground features (or background features), detects and removes false positive (or false negative) interference, and obtains more accurate segmentation results based on the initial segmentation. In addition, we design a new loss function to alleviate the insensitivity to small targets. Experimental results show that the proposed algorithm can more accurately locate pancreatic tumors of different sizes, and the Dice coefficient outperforms the existing state-of-the-art segmentation model. The code will be available at https://github.com/HeyJGJu/Pancreatic-Tumor-SEG.
Cited by 8Open access
Yangang Wang, Ningning Xi, Jiawei Hu et al. · Xibei Gongye Daxue Xuebao/Journal of Northwestern Polytechnical University · 2019
Due to the fact that there are numerous factors and parameters associated to tandem cascade, and upstream and downstream blades have intensive interference and unknown matching correlation, it is difficult to create tandem blades with high performance based on conventional optimization design method. Therefore, developing an optimization strategy that can take into account key parameters of tandem blades is a feasible way to get optimal design of tandem cascade. In the present work, the Sphere and Rastrigin test functions were adopted to compare and analyze five differential evolution modes, and the DE/r2b/2 evolution mode was then obtained, which can be applied to engineering practice. Based on the present evolutionary pattern, the combination the effects of axial overlap (AO), percent pitch (PP), camber ratio (TR), chord ratio (CR) and approximate incidence angle of the rear blade ( K b-b ) are considered comprehensively. Thus, the optimization process and strategy are designed, and the optimization objective function is determined. The optimization results show that under the condition of designed Mach number and large angle of attack (AOA), comparing with the original blades, the optimization makes the static pressure ratio of the tandem blades increased by 1.14%, and the total pressure loss coefficient decreased by 6.57%.
Cited by 0Open access
J Jagadeesan, M. Azhagiri, M. Gowtham Sethupathi · International Journal of Soft Computing and Engineering · 2025
Touchless ATMs, a new technology, offer a contactfree, hygienic, and convenient financial transaction experience. This innovative solution uses Augmented Reality (AR), Timebased One-Time Passwords (TOTP), and the HAAR Cascade Algorithm to create an interactive virtual interface, reducing physical contact and enhancing transaction security. The system uses a dual-layered authentication mechanism, utilizing facial recognition and time-based, one-time passwords (TOTP) to validate user identities and generate dynamic, session-specific codes. Financial institutions can deploy this system to upgrade their ATM networks, catering to diverse user demographics. Challenges include developing robust gesture recognition models, ensuring low latency in AR interactions, and integrating these advanced technologies into existing ATM infrastructures. However, advances in hardware and software, coupled with the decreasing cost of AR and machine learning technologies, make this solution viable and scalable.
Cited by 0Open access
M. Hoehfeld, Scott E. Fahlman · IEEE Transactions on Neural Networks · 1992
A key question in the design of specialized hardware for simulation of neural networks is whether fixed-point arithmetic of limited numerical precision can be used with existing learning algorithms. An empirical study of the effects of limited precision in cascade-correlation networks on three different learning problems is presented. It is shown that learning can fail abruptly as the precision of network weights or weight-update calculations is reduced below a certain level, typically about 13 bits including the sign. Techniques for dynamic rescaling and probabilistic rounding that allow reliable convergence down to 7 bits of precision or less, with only a small and gradual reduction in the quality of the solutions, are introduced.
Cited by 130
Unknown, Unknown, Unknown et al. · Springer Link (Chiba Institute of Technology) · 2020
\nDue to the fact that there are numerous factors and parameters associated to tandem cascade, and upstream and downstream blades have intensive interference and unknown matching correlation, it is difficult to create tandem blades with high performance based on conventional optimization design method. Therefore, developing an optimization strategy that can take into account key parameters of tandem blades is a feasible way to get optimal design of tandem cascade. In the present work, the Sphere and Rastrigin test functions were adopted to compare and analyze five differential evolution modes, and the DE/r2b/2 evolution mode was then obtained, which can be applied to engineering practice. Based on the present evolutionary pattern, the combination the effects of axial overlap (AO), percent pitch (PP), camber ratio (TR), chord ratio (CR) and approximate incidence angle of the rear blade (Kb-b) are considered comprehensively. Thus, the optimization process and strategy are designed, and the optimization objective function is determined. The optimization results show that under the condition of designed Mach number and large angle of attack (AOA), comparing with the original blades, the optimization makes the static pressure ratio of the tandem blades increased by 1.14%, and the total pressure loss coefficient decreased by 6.57%.\n
Cited by 0Open access