Tracking of a Fixed-Shape Moving Object Based on the Gradient Descent Method

Masood, Haris; Zafar, Amad; Ali, Muhammad Umair; Hussain, Tehseen; Khan, Muhammad Attique; Tariq, Usman; Damaševičius, Robertas

doi:10.3390/s22031098

Cited by 20 publications

(8 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…according to equation ( 14), the smooth model of TV can be obtained by using the gradient descent method [35],…”

Section: B Total Variationmentioning

confidence: 99%

Non-Uniform Background Noise Suppression Method Based on Improved Total Variation Model for Wide-Field Imaging System

Zhou,

Wu,

Wang

et al. 2024

IEEE Access

View full text Add to dashboard Cite

A wide-field surveillance system with a long exposure time has a stronger detectability for dim space targets. However, with the increase in exposure time and working temperature, complex non-uniform background noise containing hot pixels of the detector cannot be ignored, seriously affecting the background and imaging quality. This article studies and proposes a high-performance denoising method, which does not use any prior knowledge of the target and can automatically remove noise from the image. This method is based on an improved total variation model to remove hot pixels and other background mixed noise in widefield system images. Firstly, using the idea of the traditional local contrast method (LCM), we utilize the significant difference in grayscale values between contaminated pixels and neighboring pixels to detect impulse noise, such as the hot pixels in the image. And then, we designed an improved adaptive maximum filtering algorithm to remove unwanted contamination, which protected target information from being lost and optimized pixels that were attacked by impulse noise. Finally, the total variation algorithm is used to eliminate residual background noise, the detector's readout noise, and non-uniform response. The method proposed in this article can effectively filter out hot pixels and non-uniform background noise while preserving the details of target edges. We conducted experiments on a large number of simulated and original images. For star maps captured in long exposure mode, the method proposed in this article has obvious advantages over several competing algorithms. The experimental results show that, compared to competitive algorithms, the algorithm proposed in this article improves PSNR by at least 13.1%, SSIM by at least 0.4%, IEF by at least 5 times, and IQI by at least 9.2%. At the same time, the algorithm in this article achieved a moderate level of computation time.INDEX TERMS wide-field surveillance system; long exposure time; non-uniform correction; local contrast method; maximum filter; total variation

show abstract

“…according to equation ( 14), the smooth model of TV can be obtained by using the gradient descent method [35],…”

Section: B Total Variationmentioning

confidence: 99%

Non-Uniform Background Noise Suppression Method Based on Improved Total Variation Model for Wide-Field Imaging System

Zhou,

Wu,

Wang

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Similarly to learning rate, batch size has a significant effect on the convergence of the gradient and accuracy growth. Trials will be performed with different batch sizes to isolate the most desirable value [12].…”

Section: Tuning the Modelmentioning

confidence: 99%

Skin cancer image classification optimization through transfer learning with Tensorflow and InceptionV3

Zhao

2023

International Conference on Mechatronics Engineering and Artificial Intelligence (MEAI 2022)

View full text Add to dashboard Cite

Skin cancer is a common illness that claims thousands of lives annually in the United States alone. Accurately identifying malignant tumors is crucial to survival but can be challenging as the visual distinctions between benign and life-threatening tumors are minimal. The purpose of this project is to explore deep learning algorithms that can be trained to systematically classify skin cancer images, create a program to execute the algorithm, and exemplify an optimization process for the program that can serve as a reference for future works. The project will adopt a transfer learning algorithm based on previous studies on the subject and select a pre-trained model given practical restraints. Then a program will be coded in Python to retrieve datasets, process images, train the model, and evaluate accuracy. Finally, the algorithm will be optimized by tuning model parameters and training restraints. The experiments revealed that the algorithm was able to perform the task with an accuracy of around 70%. Model parameters such as optimizer choice and learning rate and training restraints such as batch size and epoch count have significant impacts on the training results and require precise values for maxima accuracy and minimal overfitting.

show abstract

“…Abdulaziz et al proposes to complete the extraction of feature information in a deep learning model using a high-resolution converter 9 . Haris et al proposes to detect the object of interest by means of a filter with maximum average correlation height, while the gradient descent technique is used to optimize the particle filter as a way to help the model converge quickly and improve the accuracy of target tracking 10 . The above target tracking approach uses deep learning to analyze and process all objects in the scene to filter out interested regions.…”

Section: Introductionmentioning

confidence: 99%

Tracking moving targets of interest-based deep learning

Shangguan,

Gong,

Chen

et al. 2023

5th International Conference on Information Science, Electrical, and Automation Engineering (ISEAE 2023)

View full text Add to dashboard Cite

Tracking specific target is an important research area in machine vision. When tracking specific target encounters the same local features with light and dark variations, convolutional neural networks based on deep learning require large data sets to extract target feature information. To address these problems, this paper proposes a target of interest tracking scheme based on specific object imaging. The method allows for fast capture and processing of important information about the target of interest. In addition, small data sets are used to track the target and extract the target of interest efficiently. The target of interest can also be accurately tracked when the background and light changes. The results show that our method reduces the processing of redundant information, reduces the storage overhead, and tracks the target accurately.

show abstract

Tracking of a Fixed-Shape Moving Object Based on the Gradient Descent Method

Cited by 20 publications

References 56 publications

Non-Uniform Background Noise Suppression Method Based on Improved Total Variation Model for Wide-Field Imaging System

Non-Uniform Background Noise Suppression Method Based on Improved Total Variation Model for Wide-Field Imaging System

Skin cancer image classification optimization through transfer learning with Tensorflow and InceptionV3

Tracking moving targets of interest-based deep learning

Contact Info

Product

Resources

About