Aspects Of Computer Vision Research Articles

DFAM-DETR: Deformable Feature Based Attention Mechanism DETR on Slender Object Detection

Object detection is one of the most important aspects of computer vision, and the use of CNNs for object detection has yielded substantial results in a variety of fields. However, due to the fixed sampling in standard convolution layers, it restricts receptive fields to fixed locations and limits CNNs in geometric transformations. This leads to poor performance of CNNs for slender object detection. In order to achieve better slender object detection accuracy and efficiency, this proposed detector DFAM-DETR not only can adjust the sampling points adaptively, but also enhance the ability to focus on slender object features and extract essential information from global to local on the image through an attention mechanism. This study uses slender objects images from MS-COCO dataset. The experimental results show that DFAM-DETR achieves excellent detection performance on slender objects compared to CNN and transformer-based detectors.

Temporal Pattern Attention for Multivariate Time Series of Tennis Strokes Classification.

Human Action Recognition is a challenging task used in many applications. It interacts with many aspects of Computer Vision, Machine Learning, Deep Learning and Image Processing in order to understand human behaviours as well as identify them. It makes a significant contribution to sport analysis, by indicating players' performance level and training evaluation. The main purpose of this study is to investigate how the content of three-dimensional data influences on classification accuracy of four basic tennis strokes: forehand, backhand, volley forehand, and volley backhand. An entire player's silhouette and its combination with a tennis racket were taken into consideration as input to the classifier. Three-dimensional data were recorded using the motion capture system (Vicon Oxford, UK). The Plug-in Gait model consisting of 39 retro-reflective markers was used for the player's body acquisition. A seven-marker model was created for tennis racket capturing. The racket is represented in the form of a rigid body; therefore, all points associated with it changed their coordinates simultaneously. The Attention Temporal Graph Convolutional Network was applied for these sophisticated data. The highest accuracy, up to 93%, was achieved for the data of the whole player's silhouette together with a tennis racket. The obtained results indicated that for dynamic movements, such as tennis strokes, it is necessary to analyze the position of the whole body of the player as well as the racket position.

Detection and localization of multi-scale and oriented objects using an enhanced feature refinement algorithm.

Object detection is a fundamental aspect of computer vision, with numerous generic object detectors proposed by various researchers. The proposed work presents a novel single-stage rotation detector that can detect oriented and multi-scale objects accurately from diverse scenarios. This detector addresses the challenges faced by current rotation detectors, such as the detection of arbitrary orientations, objects that are densely arranged, and the issue of loss discontinuity. First, the detector also adopts a progressive regression form (coarse-to-fine-grained approach) that uses both horizontal anchors (speed and higher recall) and rotating anchors (oriented objects) in cluttered backgrounds. Second, the proposed detector includes a feature refinement module that helps minimize the problems related to feature angulation and reduces the number of bounding boxes generated. Finally, to address the issue of loss discontinuity, the proposed detector utilizes a newly formulated adjustable loss function that can be extended to both single-stage and two-stage detectors. The proposed detector shows outstanding performance on benchmark datasets and significantly outperforms other state-of-the-art methods in terms of speed and accuracy.

ID-YOLO: Real-Time Salient Object Detection Based on the Driver’s Fixation Region

Object detection is an important task for self-driving vehicles or advanced driver assistant systems (ADASs). Additionally, visual selective attention is a crucial neural mechanism in a driver’s vision system that can rapidly filter out unnecessary visual information in a driving scene. Some existing models detect all objects in driving scenes from the aspect of computer vision. However, in a rapidly changing driving environment, detecting salient or critical objects appearing in drivers’ interested or safety-relevant areas is more useful for ADASs. In this paper, we managed to detect salient and critical objects based on drivers’ fixation regions. To this end, we built an augmented eye tracking object detection (ETOD) dataset based on driving videos with multiple drivers’ eye movement collected by Deng <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">et al.</i> Furthermore, we proposed a real-time salient object detection network named increase-decrease YOLO (ID-YOLO) to discriminate the critical objects within the drivers’ fixation region. The proposed ID-YOLO shows excellent detection of major objects that drivers are concerned about during driving. Compared with the present object detection models in autonomous and assisted driving systems, our object detection framework simulates the selective attention mechanism of drivers. Thus, it does not detect all of the objects appearing in the driving scenes but only detects the most relevant ones for driving safety. It can largely reduce the interference of irrelevant scene information, showing potential practical applications in intelligent or assisted driving systems.

Clinical Progress and Optimization of Information Processing in Artificial Visual Prostheses.

Visual prostheses, used to assist in restoring functional vision to the visually impaired, convert captured external images into corresponding electrical stimulation patterns that are stimulated by implanted microelectrodes to induce phosphenes and eventually visual perception. Detecting and providing useful visual information to the prosthesis wearer under limited artificial vision has been an important concern in the field of visual prosthesis. Along with the development of prosthetic device design and stimulus encoding methods, researchers have explored the possibility of the application of computer vision by simulating visual perception under prosthetic vision. Effective image processing in computer vision is performed to optimize artificial visual information and improve the ability to restore various important visual functions in implant recipients, allowing them to better achieve their daily demands. This paper first reviews the recent clinical implantation of different types of visual prostheses, summarizes the artificial visual perception of implant recipients, and especially focuses on its irregularities, such as dropout and distorted phosphenes. Then, the important aspects of computer vision in the optimization of visual information processing are reviewed, and the possibilities and shortcomings of these solutions are discussed. Ultimately, the development direction and emphasis issues for improving the performance of visual prosthesis devices are summarized.

Image Caption Generator

In the modern era, image captioning has become one of the most widely required tools. Moreover, there are inbuilt applications that generate and provide a caption for a certain image, all these things are done with the help of deep neural network models. The process of generating a description of an image is called image captioning. It requires recognizing the important objects, their attributes, and the relationships among the objects in an image. It generates syntactically and semantically correct sentences. In this paper, we present a deep learning model to describe images and generate captions using computer vision and machine translation. This paper aims to detect different objects found in an image, recognize the relationships between those objects and generate captions. The dataset used is Flickr8k and the programming language used was Python3, and an ML technique called Transfer Learning will be implemented with the help of the Xception model, to demonstrate the proposed experiment. This paper will also elaborate on the functions and structure of the various Neural networks involved. Generating image captions is an important aspect of Computer Vision and Natural language processing. Image caption generators can find applications in Image segmentation as used by Facebook and Google Photos, and even more so, its use can be extended to video frames. They will easily automate the job of a person who has to interpret images. Not to mention it has immense scope in helping visually impaired people.

Supervised Learning Methods for Skin Segmentation Based on Pixel Color Classification

Over the last few years, skin segmentation has been widely applied in diverse aspects of computer vision and biometric applications including face detection, face tracking, and face/hand-gesture recognition systems. Due to its importance, we observed a reawakened interest in developing skin segmentation approaches. In this paper, we offer a comparison between five major supervised learning algorithms for skin segmentation. The algorithms involved in this comparison are: Support Vector Machines (SVM), K-Nearest-Neighbors (KNN), Naive Bayes (NB), Decision Tree (DT), and Logistic Regression (LR). Various scenarios of data pre-processing are proposed including a conversion from RGB into YCbCr color space. Using YCbCr representation gave a better performance in skin/non-skin classification. Despite the settled comparison criteria, KNN was found to be the most desirable model that provides a stable performance overall the several experiments conducted.

A Comparative Study on the Center-based Iterative Hough Transform

Circle detection is one of the most critical aspects of computer vision and has been widely studied and developed in a variety of ways. The Center-based Iterative Hough Transform (CBIHT) is a method for unassisted multiple circle detection, based upon iterative uses of a center-based voting process to determine the circle’s center coordinate. This paper gives a thorough analysis of the CBIHT as well as a comparison with the Standard Hough Transform (SHT) and its well-known variants including the Generalized Hough Transform (GHT) and the Adaptive Hough Transform (AHT). When applied to synthetic and real-life circular images, our accuracy and performance comparison studies show that (i) the CBIHT is more computationally efficient than the SHT’s brute-force algorithm; (ii) the CBIHT’s center-based voting method has greater resilience to noise than the GHT and AHT’s gradient information method; and (iii) the CBIHT’s iterative process provides an adaptability and speed in unassisted multiple circle detection similar to that of the AHT; (iv) yet, the CBIHT requires no parameters for circle detection unlike the GHT and the AHT. All in all, a comparison with other methods highlights the aforementioned merit of the CBIHT, proving the CBIHT to be an excellent choice in detecting the circles with noise in real-life images.

Denoising framework based on external prior guided rotational clustering

The image denoising model based on non-local self-similarity prior (NSS) has received extensive attention in recent years because of the repeated structure of natural image patches. Similar patches collected by exploiting NSS prior are sparse, which can be used to estimate potential lowrank subspace. Meanwhile, the modelling of natural images, such as Gaussian mixture models (GMMs), has been successful in all aspects of computer vision by reducing the patterns of image patches. However, the version of its geometric transformation (e.g. rotational transformation) cannot be matched directly by using distance. How to further reduce the patterns of the patches by geometric prior and accelerate rotational matching through parallel calculation is an issue that needs to be solved. In this study, an external guided rotational matching denoising framework is proposed. The proposed framework combines non-local, sparse and low-rank image priors and we design a parallel computing scheme. They demonstrate the performance improvement of the proposed algorithm on images with strong rotational properties and the comparison with traditional state-of-the-art denoising methods. The scalability and effectiveness of the new framework are verified by simulation experiments in public and real datasets.

Guest Editorial: Image and Video Inpainting and Denoising

The papers in this special issue comprise all aspects of computer vision and pattern recognition devoted to image and video inpainting, including related tasks like denoising, debluring, sampling, super-resolutkon enhancement, restoration, hallucination, etc. The special issue was associated to the 2018 Chalearn Looking at People Satellite ECCV Workshop1 and the 2018 ChaLearn Challenges on Image and Video Inpainting.

A COMPARATIVE STUDY OF THREE CORNER FEATURE BASED MOVING OBJECT DETECTION USING AERIAL IMAGES

Corner feature based moving objects detection is an essential and fundamental research problem in the broader aspects of computer vision and pattern recognition research domain. Performance of various corner features based aerial types image processing specially for moving objects detection is still an unsolved issue due to up and down of performance which makes it difficult to choose the appropriate corner features for detection purpose. The core part mentioned in this research is to categorize significant corner characteristics of the objects using various corner features based detection methods in image extracted from aerial video. This research demonstrated three kinds of corner features, i.e. Moravec, Susan and Harris corners due to capability of these corner features to interpret high and low intensity various for aerial types of images. Standard datasets were used to evaluate each of the corner feature based detection. Based on comprehensive experimental analysis, Harris corner was observed performing efficiently comparing with Moravec and Susan corner based detection for both datasets considered by this research. Experimental results reveals the capacity of each corner characteristics based detection methodology in terms with the effectiveness using various performance metrics for moving object detection using aerial images.

An Appraisal of Content-Based Image Retrieval (CBIR) Methods

Background: Content Based Image Retrieval (CBIR) is an aspect of computer vision and image processing that finds images that are similar to a given query image in a large scale database using the visual contents of images such as colour, texture, shape, and spatial arrangement of regions of interest (ROIs) rather than manually annotated textual keywords. A CBIR system represents an image as a feature vector and measures the similarity between the image and other images in the database for the purpose of retrieving similar images with minimal human intervention. The CBIR system has been deployed in several fields such as fingerprint identification, biodiversity information systems, digital libraries, Architectural and Engineering design, crime prevention, historical research and medicine. There are several steps involved in the development of CBIR systems. Typical examples of these steps include feature extraction and selection, indexing and similarity measurement.&#x0D; Problem: However, each of these steps has its own method. Nevertheless, there is no universally acceptable method for retrieving similar images in CBIR.&#x0D; Aim: Hence, this study examines the diverse methods used in CBIR systems. This is with the aim of revealing the strengths and weakness of each of these methods.&#x0D; Methodology: Literatures that are related to the subject matter were sought in three scientific electronic databases namely CiteseerX, Science Direct and Google scholar. The Google search engine was used to search for documents and WebPages that are appropriate to the study.&#x0D; Results: The result of the study revealed that three main features are usually extracted during CBIR. These features include colour, shape and text. The study also revealed that diverse methods that can be used for extracting each of the features in CBIR. For instance, colour space, colour histogram, colour moments, geometric moment as well as colour correlogram can be used for extracting colour features. The commonly used methods for texture feature extraction include statistical, model-based, and transform-based methods while the edge method, Fourier transform and Zernike methods can be used for extracting shape features.&#x0D; Contributions: The paper highlights the benefits and challenges of diverse methods used in CBIR. This is with the aim of revealing the methods that are more efficient for CBIR.&#x0D; Conclusion: Each of the CBIR methods has their own advantages and disadvantages. However, there is a need for a further work that will validate the reliability and efficiency of each of the method.

A novel scene classification model combining ResNet based transfer learning and data augmentation with a filter

Scene classification is a significant aspect of computer vision. Convolutional neural networks (CNNs), a development of deep learning, are a well-understood tool for image classification. But training CNNs requires large-scale datasets. Transfer learning addresses this problem and produces a solution for small-scale datasets. Because scene image classification is more complex than common image classification. We propose a novel ResNet based transfer learning model utilizing multi-layer feature fusion, taking full advantage of interlayer discriminating features and fusing them for classification by softmax regression. In addition, a novel data augmentation method with a filter useful for small-scale datasets is presented. New image patches are generated by sliding block cropping of a raw image, which are then filtered to insure that the new images sufficiently represent the original categorization. Our new ResNet based transfer learning model with enhanced data augmentation is evaluated on six benchmark scene datasets (LF, OT, FP, LS, MIT67, SUN397). Extensive experimental results show that on the six datasets our method obtains better accuracy than other state-of-the-art models.

Computer vision-based phenotyping for improvement of plant productivity: a machine learning perspective.

Employing computer vision to extract useful information from images and videos is becoming a key technique for identifying phenotypic changes in plants. Here, we review the emerging aspects of computer vision for automated plant phenotyping. Recent advances in image analysis empowered by machine learning-based techniques, including convolutional neural network-based modeling, have expanded their application to assist high-throughput plant phenotyping. Combinatorial use of multiple sensors to acquire various spectra has allowed us to noninvasively obtain a series of datasets, including those related to the development and physiological responses of plants throughout their life. Automated phenotyping platforms accelerate the elucidation of gene functions associated with traits in model plants under controlled conditions. Remote sensing techniques with image collection platforms, such as unmanned vehicles and tractors, are also emerging for large-scale field phenotyping for crop breeding and precision agriculture. Computer vision-based phenotyping will play significant roles in both the nowcasting and forecasting of plant traits through modeling of genotype/phenotype relationships.

Hybrid Sensor Network-Based Indoor Surveillance System for Intrusion Detection

This paper presents a novel hybrid sensor-based intrusion detection system for low-power surveillance in an empty, sealed indoor space with or without illumination. The proposed system includes three functional steps: (i) initial detection of an intrusion event using a sound field sensor; (ii) automatic lighting control based on the detected event, and (iii) detection and tracking the intruder using an image sensor. The proposed hybrid sensor-based surveillance system uses a sound field sensor to detect an abnormal event in a very low-light or completely dark environment for 24 h a day to reduce the power consumption. After detecting the intrusion by the sound sensor, a collaborative image sensor takes over an accurate detection and tracking tasks. The proposed hybrid system can be applied to various surveillance environments such as an office room after work, empty automobile, safety room in a bank, and armory room. This paper deals with fusion of computer-aided pattern recognition and physics-based sound field analysis that reflects the symmetric aspect of computer vision and physical analysis

Triah: an intelligent guiding system for the visually impaired

Computer vision can be deemed as an emerging field which processes real world images in order to display a variety of numerical information about them. To improve the quality of the lifestyle that the visually impaired possess, we propose an assistive model which combines the various aspects of computer vision. Our proposed model aims at detecting the number of faces by using haar cascade classifiers and integrating it with Raspberry Pi. The processed images are run through the classifier and sum total is alerted through an auditory output via headphones. This model further boasts an average accuracy of 82.5%. Our model also notifies the visually impaired of the spatial orientation of the people surrounding them.

Mathematical aspects of computer vision

Mathematical aspects of computer vision

A Multi-Object Tracking Algorithm Based on Multi-Camera

As a basic aspect of computer vision, reliable tracking of multiple objects is still an open and challenging issue for both theory studies and real applications. A novel multi-object tracking algorithm based on multiple cameras is proposed in this paper. We obtain the foreground likelihood maps in each view by modeling the background using the codebook algorithm. The view-to-view homographies are computed using several landmarks on the chosen plane. Then, we achieve the location information of multi-target at chest layer and realize the tracking task. The proposed algorithm does not require detecting the vanishing points of cameras, which reduces the complexity and improves the accuracy of the algorithm. The experimental results show that our method is robust to the occlusion and could satisfy the real-time tracking requirement.

Image based modelling with VideoTrace

Image based modeling (IBM) combines aspects of computer vision, graphics and interface design. Practitioners in each field have approached IBM within the context of their own discipline, but recently systems have emerged that harness the strengths of each. In this article we discuss approaches to IBM, the development of our own system VideoTrace, and applications now and in the future.

Using a genetic algorithm to register an uncalibrated image pair to a 3D surface model

In this paper we present a successful application of genetic algorithms to the registration of uncalibrated optical images to a 3D surface model. The problem is to find the projection matrices corresponding to the images in order to project the texture on the surface as precisely as possible. Recently, we have proposed a novel method that generalises the photo-consistency approach by Clarkson et al. to the case of uncalibrated cameras by using a genetic algorithm. In previous studies we focus on the computer vision aspects of the method, while here we analyse the genetic part. In particular, we use semi-synthetic data to study the performance of different GAs and various types of selector, mutation and crossover. New experimental results on real data are also presented to demonstrate the efficiency of the method.