A System-Level Solution for Low-Power Object Detection

Multispectral systems for detection of concealed dangerous objects are becoming more popular because of their higher effectiveness compared to mono-spectral systems. So far, the problem of detecting objects hidden under clothing was considered only in the case of airports but it is becoming more important for public places like metro stations, and government buildings. Exploration of new spectral bands as well as development of technology result in introduction of new solutions – both mono and multispectral. It has been proved that objects hidden under clothing can be detected and visualized using terahertz (THz) cameras. However, passive THz cameras still offer too low image resolution for objects recognition. Limited range is another issue of passive imagers. On the other hand new infrared cameras offer sufficient parameters to detect objects covered with fabrics in some conditions, as well as high image quality and big pixel resolutions. The purpose of the studies is to investigate and compare the possibilities of using passive cameras operating in long wavelength infrared (LWIR) and THz spectral ranges for detection of concealed objects. For the purpose of investigations, commercial imagers operating in 6.5-11.7 μm and 250GHz (1.25mm) were used. In the article, we present the measurement setup and the results of measurements in various operating conditions. Theoretical studies of both spectral bands focused on detection of objects with passive imagers are also presented.

Application of Deep Learning for Object Detection

Computer vision (CV) and human-computer interaction (HCI) are essential in many technological fields. Researchers in CV are particularly interested in real-time object detection techniques, which have a wide range of applications, including inspection systems. In this study, we design and implement real-time object detection and recognition systems using the single-shoot detector (SSD) algorithm and deep learning techniques with pre-trained models. The system can detect static and moving objects in real-time and recognize the object's class. The primary goals of this research were to investigate and develop a real-time object detection system that employs deep learning and neural systems for real-time object detection and recognition. In addition, we evaluated the free available, pre-trained models with the SSD algorithm on various types of datasets to determine which models have high accuracy and speed when detecting an object. Moreover, the system is required to be operational on reasonable equipment. We tried and evaluated several deep learning structures and techniques during the coding procedure and developed and proposed a highly accurate and efficient object detection system. This system utilizes freely available datasets such as MS Common Objects in Context (COCO), PASCAL VOC, and Kitti. We evaluated our system's accuracy using various metrics such as precision and recall. The proposed system achieved a high accuracy of 97% while detecting and recognizing real-time objects.

Recent development in object detection are greatly driven by the success of region proposal approaches and region-based convolutional neural networks (R-CNNs). In this paper, we designed and implemented an object detection system using a faster-CNN method that shares full-image convolutional features with a detection network, so as to enable nearly cost-free region proposals. Development of this system is based on the previous work on Faster R-CNN. Results shows that with this method, we could achieve high accuracy while detecting objects.

Vegetation cover classification using mixed or low-resolution scalar images is challenging. Fortunately, recently deep learning object detection methods have emerged as a replacement to the conventional machine learning methods for the detection and classification of land use and land cover. This paper presents a deep learning object detection approach for land use and land cover detection using low/mixed resolution satellite images acquired from Google Earth satellite images. Google Earth images are accessible freely using the Google Earth Pro desktop application. Our dataset consists of two (02) classes (vegetation and non-vegetation) with a total of 450 labeled images captured from different parts of Pakistan. We present a comparison of the recent anchor-free object detection model YOLOX with the anchor-based object detection model YOLOR for solving real-time problems. The end-to-end differentiability, efficient GPU utilization, and absence of hand-crafted parameters make anchor-free models a compelling choice in object detection, and yet not been explored on Land cover classification using satellite images. Our experimental study shows that YOLOX delivers an overall accuracy of 83.50% on Vegetation and 86% on Non-Vegetation classes, which outperformed YOLOR by 30% on Vegetation classes and 34% on non-Vegetation classes for our dataset. We also show how an object detection system can be used for Vegetation and Non-Vegetation classification tasks, which can then be used for change monitoring and assisting in developing geographical maps using low/mixed resolution freely available satellite images.

In this paper a multiple object detection, recognition, and tracking system for unmanned aerial vehicles (UAVs) has been studied. The system can be implemented on any UAVs platform, with the main requirement being that the UAV has a suitable onboard computational unit and a camera. It is intended to be used in a maritime object tracking system framework for UAVs, which enables a UAV to perform multiobject tracking and situational awareness of the sea surface, in real time, during a UAV operation. Using machine vision to automatically detect objects in the camera's image stream combined with the UAV's navigation data, the onboard computer is able to georeference each object detection to measure the location of the detected objects in a local North‐East (NE) coordinate frame. A tracking algorithm which uses a Kalman filter and a constant velocity motion model utilizes an object's position measurements, automatically found using the object detection algorithm, to track and estimate an object's position and velocity. Furthermore, a global‐nearest‐neighbor algorithm is applied for data association. This is achieved using a measure of distance that is based not only on the physical distance between an object's estimated position and the measured position, but also how similar the objects appear in the camera image. Four field tests were conducted at sea to verify the object detection and tracking system. One of the flight tests was a two‐object tracking scenario, which is also used in three scenarios with an additional two simulated objects. The tracking results demonstrate the effectiveness of using visual recognition for data association to avoid interchanging the two estimated object trajectories. Furthermore, real‐time computations performed on the gathered data show that the system is able to automatically detect and track the position and velocity of a boat. Given that the system had at least 100 georeferenced measurements of the boat's position, the position was estimated and tracked with an accuracy of 5–15 m from 400 m altitude while the boat was in the camera's field of view (FOV). The estimated speed and course would also converge to the object's true trajectories (measured by Global Positioning System, GPS) for the tested scenarios. This enables the system to track boats while they are outside the FOV of the camera for extended periods of time, with tracking results showing a drift in the boat's position estimate down to 1–5 m/min outside of the FOV of the camera.

Detecting objects in aerial images is challenging for at least two reasons: (1) target objects like pedestrians are very small in pixels, making them hardly distinguished from surrounding background; and (2) targets are in general sparsely and non-uniformly distributed, making the detection very inefficient. In this paper, we address both issues inspired by observing that these targets are often clustered. In particular, we propose a Clustered Detection (ClusDet) network that unifies object clustering and detection in an end-to-end framework. The key components in ClusDet include a cluster proposal sub-network (CPNet), a scale estimation sub-network (ScaleNet), and a dedicated detection network (DetecNet). Given an input image, CPNet produces object cluster regions and ScaleNet estimates object scales for these regions. Then, each scale-normalized cluster region is fed into DetecNet for object detection. ClusDet has several advantages over previous solutions: (1) it greatly reduces the number of chips for final object detection and hence achieves high running time efficiency, (2) the cluster-based scale estimation is more accurate than previously used single-object based ones, hence effectively improves the detection for small objects, and (3) the final DetecNet is dedicated for clustered regions and implicitly models the prior context information so as to boost detection accuracy. The proposed method is tested on three popular aerial image datasets including VisDrone, UAVDT and DOTA. In all experiments, ClusDet achieves promising performance in comparison with state-of-the-art detectors.

The application of deep learning technology has increased rapidly in recent years. Technologies in deep learning increasingly emulate natural human abilities, such as knowledge learning, problem-solving, and decision-making. In general, deep learning can carry out self-training without repetitive programming by humans. Convolutional neural networks (CNNs) are deep learning algorithms commonly used in wide applications. CNN is often used for image classification, segmentation, object detection, video processing, natural language processing, and speech recognition. CNN has four layers: convolution layer, pooling layer, fully connected layer, and non-linear layer. The convolutional layer uses kernel filters to calculate the convolution of the input image by extracting the fundamental features. The pooling layer combines two successive convolutional layers. The third layer is the fully connected layer, commonly called the convolutional output layer. The activation function defines the output of a neural network, such as 'yes' or 'no'. The most common and popular CNN activation functions are Sigmoid, Tanh, ReLU, Leaky ReLU, Noisy ReLU, and Parametric Linear Units. The organization and function of the visual cortex greatly influence CNN architecture because it is designed to resemble the neuronal connections in the human brain. Some of the popular CNN architectures are LeNet, AlexNet and VGGNet.

Object detection systems are central to the autonomy and safety of intelligent transportation systems. Yet, the accuracy of object detection models can suffer under environmental noise or adverse weather. This paper tested the robustness of four object detection architectures: YOLOv5s, YOLOv8m, YOLOv10n, and Faster R-CNN, to visual degradation (real-world weather and artificial noise). We utilize the DAWN dataset, a benchmark of 1,000 high-resolution traffic images captured under fog, rain, snow, and sandstorms, with further augmentations of Gaussian noise, salt-and-pepper noise, blurriness, and overlays applied with artificial fog. We standardized all annotations to YOLO and COCO annotation formats for multi-framework interoperability. Our quantitative analysis used mAP@0.5, mAP@0.5:0.95, Precision, and Recall to compare the models, alongside some qualitative analysis through visual overlays and plotting training loss. The findings of our analysis showed YOLOv8m achieved the highest baseline accuracy on clean data, while Faster R-CNN proved resilient in noisy environments. YOLOv10n achieves a good trade-off between efficient and robust detection. The results of this study highlight the necessity of adaptive training pipelines and environment-aware benchmarks to enhance the real-world reliability of vision-based detection systems.

In industrial visual inspection, foreign matters are mostly fractal objects. Detailed detection of fractal objects is difficult but necessary because better decision-making relies on more detailed and more comprehensive detection information. This presents a challenge for industrial applications. To solve this problem, we proposed a holistic approach to fractal object detection based on a multi-head model. We proposed the IWS (Information Watch and Study) module to provide enhancement learning capabilities for object information. It increases the detection dimension of the object and can perform more detailed detection. In order to realize the portability of the IWS module, it can be easily and quickly deployed to the existing advanced object detection model to achieve end-to-end detection. We proposed the FGI (Fine-Grained Information) Head, which is used to extract more comprehensive feature vectors from the original base model. We proposed the WST (Watch and Study Tactic) Learner for object information processing and adaptive learning of class cluster centers. Using the MRD (Multi-task Result Determination) strategy to combine the classification results and IWS results, the final detection results are derived. In the experiment, the IWS and MRD were mounted on three different models of the YOLO series. The experimental results show that YOLO+IWS has good foreign object detection capabilities to meet the needs of industrial visual inspection. Moreover, for the detailed detection ability of fractal objects, YOLO+IWS is better than the other 11 competing methods. We designed a new evaluation index and an adjustment mechanism of class learning weights to make better judgments and more balanced learning. Not only that, we applied YOLO+IWS to form a brand new object detection system.

Abstract—Object detection is a computer vision technique that locates objects in images or videos by creating bounding boxes around them. In this paper, we propose a model based on object detection using deep learning technologies along with text to speech conversion.An object detection system uses a deep learningmodel to detect objects using YOLO (You Only Look Once) and text-to- speech (TTS) to synthesize a voice announcement about each object. The system we used is built using python OpenCV tool and Google text to speech (gTTS) is used to convert text into audio segment. First variations of YOLO algorithm are compared and then the best one is used according to result we get it by training it on COCO dataset. After the object is detected, the name of the detected object is displayed then the voice output is generated by using Google Text To Speech(gTTS) module. The contribution we make is to present a visual substitution system that uses features extraction and matching to recognize objects with a voice feedback. Index Terms—Object Detection, YOLO, Open CV, python, Google Text To Speech

With the rapid development of autonomous driving technology, traditional object detection algorithms often face challenges such as the trade-off between model parameters and recognition accuracy, insufficient detection accuracy, and slow detection speed when dealing with vehicle and pedestrian detection tasks in complex scenarios. To address these challenges, we propose a novel lightweight real-time driving scene detection network, FEEP-YOLO. Based on the YOLOv8 algorithm, our network not only maintains high detection accuracy but also achieves faster training speed and fewer model parameters. FEEP-YOLO employs a lightweight network, Faster Net, to significantly reduce model parameters, thereby accelerating detection speed and better meeting real-time requirements. To further improve detection accuracy, we introduce an EMA attention module into the network and optimize the bounding box loss function, adopting the EIOU loss function to enhance target features and reduce background interference. Additionally, to detect small objects more accurately, we add feature layers rich in semantic information to the detection end. Experimental results demonstrate that FEEP-YOLO outperforms YOLOv8 in terms of both detection speed and accuracy on the KITTI and SODA datasets, while significantly reducing the number of model parameters and model size. FEEP-YOLO achieves high detection speed while improving detection accuracy and reducing computational resource consumption, meeting the practical application requirements under limited computing power, memory space and power consumption. The key contribution of FEEP-YOLO lies in its ability to balance accuracy speed, and efficiency, making it a robust solution for real-time autonomous driving applications in complex environments.

UAV imagery is widely used in areas like traffic safety, disaster rescue, and airspace management, due to its small size and low cost. However, it poses unique challenges for object detection due to small objects, complex backgrounds, and noise interference. To tackle these challenges, we propose YOLO-Air, a novel small object detection network designed specifically for UAV imagery. We propose SECAConv (Squeeze-Excitation Convolution with Attention), which enhances the feature representation of small objects through dynamic weight allocation and channel attention mechanisms. Additionally, we design the novel AeroFPN (Aerial Feature Pyramid Network) to optimize feature transmission by alleviating shallow feature loss through the inclusion of the xsmall detection head. Furthermore, we develop ASFM (Adaptive Scale Fusion Module), which suppresses background noise interference through effective multi-scale feature fusion and adaptive channel attention mechanisms, thereby improving the network’s ability to detect small objects. Experimental results demonstrate that YOLO-Air achieves significant accuracy improvements on both the VisDrone-DET2019 and AI-TOD datasets. Compared to the baseline YOLOv8n, YOLO-Air improved <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">mAP</i><sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">50</sub> from 41.2% to 44.5% on the VisDrone-DET2019 dataset, and from 44.9% to 47.5% on the AI-TODdataset, while maintaining computational efficiency. These results validateYOLO-Air as an effective solution for small object detection in UAV aerial imagery.

Robust and accurate sensing is of critical importance for advancing autonomous automotive systems. The need to acquire situational awareness in complex urban conditions using sensors such as radar has motivated research on power and latency-efficient signal acquisition methods. In this paper, we present an end-to-end signal processing pipeline, capable of operating in extreme weather conditions, that relies on sub-sampled radar data to perform object detection in vehicular settings. The results of the object detection are further utilized to sub-sample forthcoming radar data, which stands in contrast to prior work where the sub-sampling relies on image information. We show robust detection based on radar data reconstructed using 20 % of samples under extreme weather conditions such as snow or fog, and on low-illuminated nights. Additionally, we generate 20% sampled radar data in a fine-tuning set and show 1.1 % gain in AP50 across scenes and 3% AP50 gain in motorway condition.

Portable assistive technology systems are developed to enhance the capability of persons with disabilities. One of the most important senses for humans is vision. Vision is the most important sense which helps us in understanding the perception of the surrounding environment. Visually impaired persons face a lot of difficulty in understanding the perception around them, particularly in outdoor environment where objects are continuously changing and moving from one place to another. Object detection solutions would greatly assist visually impaired persons in avoiding from the barriers which they face in their daily routine life. The aim of the object detection system is to provide a simple, user-friendly, handy, economical and efficient solution for the visually impaired persons. Motive of this system is to develop a solution that detects the objects present using camera, as the input device in real time and communicate the same to the user using smartphone through headphones. The system would be using an audio device such as speakers or headphones in providing the information about objects to assist the visually impaired persons. The proposed system helps in identifying and avoiding the objects both in outdoor and indoor environments that affect day to day life activities and occupational performance of the visually impaired persons. The information about the objects in surrounding environment would be very much helpful to the visually impaired persons in their daily life.KeywordsObject detection systemVisually impairedBlindnessAssistive systemComputer visionImage recognition