Object Detection Area Research Articles

The detection and classification of acute myeloid leukaemia blood cell images based on different YOLO approaches

Medical image examination with a deep learning approach is greatly beneficial in the healthcare industry for faster diagnosis and disease monitoring. One of the popular deep learning algorithms such as you only look once (YOLO) developed for object detection is a successful state-ofthe-art algorithm in real-time object detection systems. Although YOLO is continuously improving in the object detection area, there are still questions about how different YOLO versions compare in terms of performance. We utilize eight YOLO versions to classify acute myeloid leukaemia (AML) blood cells in image examinations. We also acquired the publicly available AML dataset from the cancer imaging archive (TCIA) which consists of expert-labeled single cell images. Data augmentation techniques are additionally applied to enhance and balance the training images in the dataset. The overall results indicated that eight types of YOLO approaches have outstanding performances of more than 90% in precision and sensitivity. In comparison, YOLOv4-tiny has a more reliable performance than the other seven approaches. Consistently, the YOLOv4-tiny also achieved the highest AUC score. Therefore, this work can potentially provide a beneficial digital rapid tool in the screening and evaluation of numerous haematological disorders.

Optimal Design for Modulated Binary Sensors

A paramount factor limiting the applications of binary sensors is these senors’ on-off property outputting binary digits of “0” or “1”. To overcome this limitation, modulators or obscurants are added to enhance the sensing ability of binary sensors and render them usable in applications such as multi-target tracking and human activity recognition. Obscurants segment the field of interest into subregions and distinguish each subregion by a list of sensor states called signatures. This paper studies two placement scenarios in a two-dimensional planar graph. In the first scenario, we prove upper and lower bounds on the maximum number of achievable signatures. In the second scenario, starting from the placement of sensors and obscurants in which the maximum number of signatures is achievable, we propose a novel mathematical model based on four main metrics: the object space size, the sensor space size, the obscurant space size, and the sizes of individual obscurants. We find the minimum and maximum radiuses that bound the object detection area given the number and sizes of sensors and obscurants. We derive the obscurant space size as a function of the object space size and the size of individual obscurants. We also provide a linear relationship formula between the obscurant space radius, the sensor space radius, and the obscurants’ radiuses and conduct modeling experiments to study the relationship between these metrics. Finally, we deduct an explicit formula for the maximum obscurant space size for the sensor space and the individual obscurant sizes.

ObserveNet Control: A Vision-Dynamics Learning Approach to Predictive Control in Autonomous Vehicles

A key component in autonomous driving is the ability of the self-driving car to understand, track and predict the dynamics of the surrounding environment. Although there is significant work in the area of object detection, tracking and observations prediction, there is no prior work demonstrating that raw observations prediction can be used for motion planning and control. In this paper, we propose ObserveNet Control, which is a vision-dynamics approach to the predictive control problem of autonomous vehicles. Our method is composed of a: i) deep neural network able to confidently predict future sensory data on a time horizon of up to 10s and ii) a temporal planner designed to compute a safe vehicle state trajectory based on the predicted sensory data. Given the vehicle's historical state and sensing data in the form of Lidar point clouds, the method aims to learn the dynamics of the observed driving environment in a self-supervised manner, without the need to manually specify training labels. The experiments are performed both in simulation and real-life, using CARLA and RovisLab's AMTU mobile platform as a 1:4 scaled model of a car. We evaluate the capabilities of ObserveNet Control in aggressive driving contexts, such as overtaking maneuvers or side cut-off situations, while comparing the results with a baseline Dynamic Window Approach (DWA) and two state-of-the-art imitation learning systems, that is, Learning by Cheating (LBC) and World on Rails (WOR).

A Systematic Review of Algorithms in People Images Detection Based on Artificial Vision Techniques for Energy Management in Air Conditioners

Over the years, the development of artificial intelligence has influenced the fact that the algorithms applied in video devices are renewed every day in their object detection area, such as pattern recognition for detecting an object, image, person. This research aims to identify the algorithms for detecting people's image through artificial vision, and its application focused on energy management in air conditioners. The following research questions were established: Q1: How many studies refer to algorithms based on detecting a person's image? Q2: How many studies refer to energy management in air conditioners based on artificial vision? Q3: Are there artificial vision techniques for the automatic turning on and off a device? Q4: What types of algorithms in detecting a person's image based on artificial vision exist today? The search for information was based on research criteria related to the topic developed in four virtual libraries. This research criterion includes 213 relevant studies, including 46 that were of interest for developing the research. This research's results determine that the viola-jones algorithm shows greater effectiveness with 91.5% in relation to the other algorithms, followed by the HOG algorithm with 90.29% effectiveness. Considering the different efficiency parameters established in this study, we can conclude that this algorithm can be used in various applications, security, energy management, video surveillance and environment.

Trident SSD: A Trident Single-Shot Multibox Object Detector with Deconvolution

Scale variation is one of the key challenges in the object detection area, which limits the precision of detection methods like Single shot multibox detector (SSD). This paper proposes a detection method based on SSD, which focuses on handling scale variation and better detection performance of small objects. Our method, called TridentSSD, have an architecture with three branches, which are respectively responsible for detecting different scales of objects, solving the problem of scale variation while training. Then we augment small object branch with deconvolution module and feature fusion methods to improve precision, especially for small object detection. During training, we modify the original matching rules to generate training samples. Consequently, we can use objects of different scales to train the corresponding branches respectively. Finally, Experiments have done on both PASCAL VOC2007 and VOC2012 datasets. Results show that, with an input size of 300×300, our TridentSSD achieves better performance compared to the benchmark method SSD.

The Fast Integration of a Rotated Rectangle Applied to the Rotated Haar-Like Features for Rotated Objects Detection

In the area of Object Detection, the most important step is the extraction of object features. One of the most used approaches is HaarLike features and the Integral Image technique to integrate them. The Integral Image technique, used by Viola and Jones, is generally used to calculate the integral of a rectangular filter in an input picture. This filter is a rectilinear rectangle. We propose a method to integrate a rotated one by any angle of rotation inside an image based on the Bresenham algorithm of drawing a segment. We use some pixels – called key points - that forms the four segments of a rotated rectangle, to calculate its Integral Image. Our method focuses on three essential tasks; the first is to determine the rule for drawing a segment (SDR), the second is to identify all the key points of the rectangle r, and the third is to calculate the integral image. The speed of this method depends on the size and angle of rotation of the rectangle. To demonstrate the efficiency of our idea, we applied it to the rotated Haar-like features that we proposed in a later work [12], which had as objectives the improvement of the Viola and Jones algorithm to detect the rotated faces in a given image. We performed tests on more widespread databases of images, which showed that the application of this technique to rotated Haar-Like features improves the performance of object detectors, in general, and faces in particular.

A Visual Object Tracking Algorithm Based on Improved TLD

Visual object tracking is an important research topic in the field of computer vision. Tracking–learning–detection (TLD) decomposes the tracking problem into three modules—tracking, learning, and detection—which provides effective ideas for solving the tracking problem. In order to improve the tracking performance of the TLD tracker, three improvements are proposed in this paper. The built-in tracking module is replaced with a kernelized correlation filter (KCF) algorithm based on the histogram of oriented gradient (HOG) descriptor in the tracking module. Failure detection is added for the response of KCF to identify whether KCF loses the target. A more specific detection area of the detection module is obtained through the estimated location provided by the tracking module. With the above operations, the scanning area of object detection is reduced, and a full frame search is required in the detection module if objects fails to be tracked in the tracking module. Comparative experiments were conducted on the object tracking benchmark (OTB) and the results showed that the tracking speed and accuracy was improved. Further, the TLD tracker performed better in different challenging scenarios with the proposed method, such as motion blur, occlusion, and environmental changes. Moreover, the improved TLD achieved outstanding tracking performance compared with common tracking algorithms.

DATA PROCESSING AND RECORDING USING A VERSATILE MULTI-SENSOR VEHICLE

Abstract. In this paper we present a versatile multi-sensor vehicle which is used in several research projects. The vehicle is equipped with various sensors in order to cover the needs of different research projects in the area of object detection and tracking, mobile mapping and change detection. We show an example for the capabilities of this vehicle by presenting camera- and LiDAR-based pedestrian detection methods. Besides this specific use case, we provide a more general in-depth description of the vehicle’s hard- and software design and its data-processing capabilities. The vehicle can be used as a sensor carrier for mobile mapping, but it also offers hardware and software components to allow for an adaptable onboard processing. This enables the development and testing of methods related to real-time applications or high-level driver assistance functions. The vehicle’s hardware and software layout result from several years of experience, and our lessons learned can help other researchers set up their own experimental platform.

Hazardous Object Detection by Using Kinect Sensor in a Handle-Type Electric Wheelchair.

To ensure the safety of a handle-type electric wheelchair (hereinafter, electric wheelchair), this paper describes the applicability of using a Kinect sensor. Ensuring the mobility of elderly people is a particularly important issue to be resolved. An electric wheelchair is useful as a means of transportation for elderly people. Considering that the users of electric wheelchairs are elderly people, it is important to ensure the safety of electric wheelchairs at night. To ensure the safety of an electric wheelchair at night, we constructed a hazardous object detection system using commercially available and inexpensive Kinect sensors and examined the applicability of the system. We examined warning timing with consideration to the cognition, judgment, and operation time of elderly people. Based on this, a hazardous object detection area was determined. Furthermore, the detection of static and dynamic hazardous objects was carried out at night and the results showed that the system was able to detect with high accuracy. We also conducted experiments related to dynamic hazardous object detection during daytime. From the above, it showed that the system could be applicable to ensuring the safety of the handle-type electric wheelchair.

A review and an approach for object detection in images

An object detection system finds objects of the real world present either in a digital image or a video, where the object can belong to any class of objects namely humans, cars, etc. In order to detect an object in an image or a video the system needs to have a few components in order to complete the task of detecting an object, they are a model database, a feature detector, a hypothesiser and a hypothesiser verifier. This paper presents a review of the various techniques that are used to detect an object, localise an object, categorise an object, extract features, appearance information, and many more, in images and videos. The comments are drawn based on the studied literature and key issues are also identified relevant to the object detection. Information about the source codes and online datasets is provided to facilitate the new researcher in object detection area. An idea about the possible solution for the multi class object detection is also presented. This paper is suitable for the researchers who are the beginners in this domain.

Occlusion Detection and Handling: A Review

Object tracking and detection is a classical research area in the field of computer vision from decades. Numerous kinds of applications are dependent on the area of object detection, such as advance driving assistance system, traffic surveillance, scene understanding, autonomous navigation etc. Many challenges still exist while detecting an object such as illusion, low visibility, cast shadows and most importantly occlusions of object. Occlusions occur under two categories, firstly its, self‐occlusion which means that, from a certain viewpoint, one part of an object is occluded by another part. Secondly, its inter-object occlusion which means when two objects being tracked occlude each other. We will review various occlusion handling methods that involved single and multiple cameras according to their application. In short, the objective of this paper is to deliberate in detail the problem of occlusion in object tracking and provide a concise review for the problem of occlusion handling under different categories and identify new trends.

Study on a Rapid Real-Time Feature Extraction Algorithm

A fast feature extraction algorithm is presented in this paper based on the color and point feature, With the aim of required feature points, the location of the object includes object region detection and feature point location. The area of object detection is used to look for the centre of gravity point from the scaling image. The feature point location is based on the object region detection, cuts a picture from the image, and extracts speeded up robust feature (SURF) points of the target within the cut picture. The target position is calculated according to the value of the feature points, it provides a basis for object tracking. The experimental results verify the effectiveness of the proposed method.

A Novel Human Detection Algorithm Based on Foreground Segmentation

In computer vision applications, human detection occupies an important position. HOG (Histograms of Oriented Gradient) is a classical algorithm which was used in the area of object detection. But the complex background would greatly affect the test accuracy when taking HOG as a human characteristic for human detection. In order to improve the accuracy of human detection, this paper applied a new algorithm which was based on foreground segmentation. We could get each closed region by Oriented Watershed Transform and Ultrametric Contour Map, then the foreground and the background could be distinguished. Finally we removed the background and calculated the foreground characteristic. The experimental results show that this approach was effective in improving detection accuracy. c 2013 Binary Information Press a Textile Bioengineering and Informatics Society September 2013.