Object Detection with Deep Learning

Dear Colleagues,

Object detection is one of the most important fundamental branches within the realm of computer vision. It has been widely applied in numerous real-world applications, such as security surveillance, autonomous driving, remote sensing scene analysis, robotic vision, and so on. The accurate localization and classification of objects within complex and diverse scenes represent critical challenges in object detection, limiting the development of object detection technology. With the rapid development of deep learning networks for object detection tasks, the performance of object detectors has achieved remarkable performance improvements and opened up new avenues for research. Deep-learning-based approaches have demonstrated significant success in addressing this challenge; consequently, convolutional neural networks (CNNs) have allowed us to understand rich representations of objects from raw image data. This Special Issue aims to furnish a thorough exploration of recent advancements and emerging trends in the domain of deep learning applied to object detection, including the development of novel architecture design, attention mechanisms for feature extraction, training methodologies, model compression, and various specific applications.

Topic areas include, but are not limited to, the following:

• Few-shot/zero-shot object detection;
• Weak/semi/unsupervised object detection;
• Model compression for object detectors;
• Small object detection;
• Rotated object detection

Dr. Shun Zhang
Dr. Feng Gao
Dr. Mingyang Ma
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Future Internet is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• object detection
• deep learning
• few-shot object detection
• model compression
• rotated object detection

Title: Analysis and Review of Deep Learning Models on Medical Images for Disease Diagnosis
Authors: Hamid GholamHossieni (1,*); Dilhani Thakshila Dodangoda (1); Ranpreet Kaur (2)
Affiliation: 1 School of Engineering, Computer and Mathematical Sciences, AUT University, Auckland, New Zealand; [email protected]; [email protected] 2 Software Engineering, Media Design School, Auckland, New Zealand; [email protected] * Correspondence: [email protected]; Tel.: (+649 921 9999-5455)
Abstract: Object detection in medical images is crucial for early diagnosis and timely intervention, contributing significantly to patient care. Developing robust object-detection methods that ensure efficient and precise detection while preserving image integrity is essential. Deep learning techniques have emerged as a powerful solution for object detection and classification in various applications, outperforming traditional methods due to their advanced feature learning and training capabilities. We reviewed state-of-the-art deep learning networks for medical image segmentation and object detection. Furthermore, we conducted a comparative analysis using these networks and our proposed deep learning algorithm utilizing skin images from the ISIC dataset. The evaluation considered metrics such as processing time, complexity, hardware implementation, and accuracy. Results showed that incorporating deep learning algorithms into computer-aided diagnostic (CAD) systems enhances the performance, accuracy, and precision of object detection, thereby minimizing errors and improving diagnostic outcomes. The findings highlight the potential for achieving higher accuracy in skin mole segmentation, demonstrating the effectiveness of deep learning in medical image classification and diagnosis. This underscores the promising role of deep learning networks in improving patient care and clinical decision-making.