Scale Proportionate Histograms of Oriented Gradients for Object Detection in Co-Registered Visual and Range Data

In complex scene, considering traditional object detection methods based on feature points have exposed many problems, such as undetected points, low detected ratio and cannot well process object occlusion and scaling situation, this paper proposes a detection method which based on a deformable part model. The method uses histogram of oriented gradient (HOG) feature as the object description, and the deformable part model includes a global template and several high-resolution templates. And the method uses the support vector machine (SVM) training the object model. In the learning process, after the HOG feature extracted, the method modifies the HOG feature, and then uses the principal component analysis (PCA) method reducing feature dimensions to avoid over-learning, and improve the detection rate in the detection process. The experiment results shows that the method proposed can better process object occlusion or scaled situation, and there’s also an improvement in detection ratio.

A fusion of dolphin swarm optimization and improved sine cosine algorithm for automatic detection and classification of objects from surveillance videos

Object detection in images is a computing demanding task which usually needs to deal with the detection of different classes of objects, and thus requiring variations and adaptations easily provided by software solutions. Object detection algorithms are being part of real-time smarter embedded systems, such as automotive, medical, robotics and security systems. In most embedded systems, efficient implementations of object oriented algorithms need to provide high performance, low power consumption, and programmability to allow greater development flexibility. The Histogram of Oriented Gradients (HOG) is one of the most widely used algorithms for object detection in images. In this paper, we show our work towards mapping the HOG algorithm to an FPGA-based system consisting of multiple Nios II softcore processors and bearing in mind high-performance and programmability issues. We show how to reduce 19x the algorithms execution time by source to source transformations and specially avoiding redundant processing. Furthermore, we show how the use of pipelining processing using three Nios II processors allows a speedup of 49x compared to the embedded baseline application.

Object detection and localization in an image can be achieved by representing an object as a Histogram of Oriented Gradients (HOG). HOGs have proven to be robust object descriptors. However, to achieve accurate object localization, one must take a sliding window approach and evaluate the similarity of the descriptor over all possible windows in an image. In case that search should also be scale and rotation invariant, the exhaustive consideration of all possible HOG transformations makes the method impractical due to its computational complexity. In this work, we first propose a variant of an existing rotation invariant HOG-like descriptor. We then formulate object detection and localization as an optimization problem that is solved using the Particle Swarm Optimization (PSO) method. A series of experiments demonstrates that the proposed approach results in very large performance gains without sacrificing object detection and localization accuracy.KeywordsParticle Swarm OptimizationObject DetectionLocalization AccuracyObject LocalizationReference ObjectThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Fire is a disaster caused by accidental fire, unexpected or unwanted, difficult to control, and detrimental. Many factors cause fires. Humans may directly or indirectly cause these factors, or they may also be caused by nature. Total 913 residential fires were recorded in Indonesia from 2013 to 2015. There are still many incidents of a fire occurring. This research is carried out to reduce the incident to detect fire objects by applying the histogram of oriented gradient (HOG) method and then classified using the support vector machine (SVM) method. The proposal is expected to help in the early identification of fires to reduce the number of fires that occur. This study has several stages: inserting images, converting images to grayscale, thresholding methods, object detection with HOG, and classification using SVM. The system implementation is carried out using the Python programming language with the Django web framework. The final result will be tested using a confusion matrix with four matrices: accuracy, precision, recall, and F1-Score. The results showed that the HOG and SVM methods using the RBF kernel and the sharing of training-test data 70:30 obtained accuracy of 88.33%, precision of 85.71 %, recall of 88.89%, and F1-Score of 87. 27%.

In this work, we propose a novel technique to generate shapes from point cloud data. A point cloud can be viewed as samples from a distribution of 3D points whose density is concentrated near the surface of the shape. Point cloud generation thus amounts to moving randomly sampled points to high-density areas. We generate point clouds by performing stochastic gradient ascent on an unnormalized probability density, thereby moving sampled points toward the high-likelihood regions. Our model directly predicts the gradient of the log density field and can be trained with a simple objective adapted from score-based generative models. We show that our method can reach state-of-the-art performance for point cloud auto-encoding and generation, while also allowing for extraction of a high-quality implicit surface. Code is available at https://github.com/RuojinCai/ShapeGF .

Unmanned aerial vehicles (UAVs) or drones are lightly weighted platforms with numerous advantages and can acquire very high-resolution data rapidly with less expenditure. However, the processing of very high-resolution data acquired through the sensors mounted on the UAV platform is a major challenge, and the commercial UAV data processing software is expensive. These software packages require high-configuration hardware and follow the general workflow of conventional photogrammetric approach for image alignment, point cloud generation, mesh generation, DSM, and ortho-mosaic generation. The integration of computer vision algorithms into digital photogrammetric techniques has made it possible to process the high-resolution datasets quickly. A wide variety of open-source software utilities are available that can perform these tasks individually; however, an integrated solution is always desirable. An attempt is made to develop an open-source tool for photogrammetric processing of UAV data using open-source libraries in Python. The software provides a complete workflow for photogrammetric processing of UAV data ranging from camera calibration, point cloud generation from selective frames based on B/H ratio, direct georeferencing of generated point cloud and digital elevation model (DEM) generation. The developed software is compatible with both 32-bit and 64-bit operating systems and low configuration hardware.

: We investigate an object detection system that uses both image and three-dimensional (3-D) point cloud data captured from the low-cost Microsoft Kinect vision sensor. The system works in three parts: image and point cloud data are fed into two components; the point cloud is segmented into hypothesized objects and the image region for those objects are extracted; and finally, a histogram of oriented gradient (HOG) descriptors are used for detection using a sliding window scheme. We evaluate this system by detecting backpacks on a challenging set of capture sequences in an indoor office environment with encouraging results.

In the field of 3D point cloud generation, numerous 3D generative models have demonstrated the ability to generate diverse and realistic 3D shapes. However, the majority of these approaches struggle to generate controllable 3D point cloud shapes that meet user-specific requirements, hindering the large-scale application of 3D point cloud generation. To address the challenge of lacking control in 3D point cloud generation, we are the first to propose controlling the generation of point clouds by shape structures that comprise part existences and part adjacency relationships. We manually annotate the adjacency relationships between the segmented parts of point cloud shapes, thereby constructing a StructureGraph representation. Based on this StructureGraph representation, we introduce StrucADT, a novel structure-controllable point cloud generation model, which consists of StructureGraphNet module to extract structure-aware latent features, cCNF Prior module to learn the distribution of the latent features controlled by the part adjacency, and Diffusion Transformer module conditioned on the latent features and part adjacency to generate structure-consistent point cloud shapes. Experimental results demonstrate that our structure-controllable 3D point cloud generation method produces high-quality and diverse point cloud shapes, enabling the generation of controllable point clouds based on user-specified shape structures and achieving state-of-the-art performance in controllable point cloud generation on the ShapeNet dataset.

In order to detect noncooperative objects with unknown structures, template based matching approaches are generally adopted. They rely on a large set of manually-selected templates and slide them over the image to determine the potential locations of objects. The process is exhaustive and computationally inefficient. In this paper, we propose a novel object detection algorithm using improved features of histogram of oriented gradients (HOG) to reduce the search region of potential objects regardless of their prior information. Firstly, we improve the HOG descriptor to make it more discriminative. The capability of detecting objects comes from positive and negative features of the training dataset. Then, the cascaded support vector machine is used to train the model, aiming at selecting proposals with higher scores at each scale and aspect ratio. Lastly, the best proposal over all scales is chosen as the object detection region. Further experiments demonstrate that our method improves favorably the detection rate on VOC 2007 and achieves satisfying performance in satellite bracket detection.

This paper presents a novel approach to construct spatially-referenced, multidimensional virtual city models from remotely generated point clouds for areas that lack reliable geographical reference data. A multidimensional point cloud is an unstructured array of single, irregular points in a spatial 3D coordinate system plus time stamp. If geospatial reference points are available, a point cloud is geo-referenced. Geo-referenced point clouds contain a high-precision reference dataset. Point clouds can be utilised in a variety of applications. They are particularly suitable for the representation of surfaces, structures, terrain and objects. Point clouds are used here to generate a virtual 3D city model representing the complex, granular cityscape of Jerusalem and its centre, the Old City. The generation of point clouds is based on two data acquisition methods: active data capture by laser scanning and passive data collection by photogrammetric methods. In our case, very high-resolution stereo imagery in visible light and near infrared bands have been systematically acquired an aerial flight campaign. The spatio-temporal data gathered necessitate further processing to extract the geographical reference and semantic features required in a specific resolution and scale. An insight is given into the processing of an unstructured point cloud to extract and classify the 3D urban fabric and reconstruct its objects. Eventually, customised, precise and up-to-date geographical datasets can be made available for a defined region at a defined resolution and scale.

The article explores modern neural network architectures for the automatic detection and recognition of marine surface objects and obstacles of given classes throughout the full image area, applicable for execution in real or near real time on an optoelectronic vision system to au-tomate and improve the safety of civil marine navigation. A formal statement of the problem of automatic detection of objects on images is given. The state-of-the-art algorithms for detecting objects in images based on use of artificial convolutional neural networks were reviewed, their comparison was made and a reasonable choice was made in favor of the most efficient neuralnetwork architecture in terms of computational complexity to recognition accuracy. The subject area is studied, as well as publicly available databases of surface objects suitable for use in the training of algorithms using artificial neural networks. The article concluded that there is insuffi-cient labeled data for training neural network algorithms, as a result of which the authors inde-pendently collected research images and video sequences, prepared and labeled the collected data containing surface marine objects and other obstacles that represent a navigation hazard for ships. Based on the selected neural network architecture, a new neural network algorithm for automatic full-frame detection and recognition of surface objects was developed, and an artificial neural network was trained using the prepared database of images of typical objects. The resulting algorithm was tested by the authors on a validation data set, the quality of its work was estimated using various metrics, and the algorithm’s performance was measured. Conclusions are made about the necessity to expand the collected database of images of typical marine objects, further steps are proposed to improve the accuracy of the developed software and algorithmic complex and its implementation to be used in a marine optoelectronic machine vision system for automa-tion and improving the safety of civil navigation.

Many possible use-cases for deep-learning-based 3D object recognition in logistic environments have been proposed [1]. Nonetheless, the applicability of this technology in a logistic context has yet to be explored. Additionally, no dataset regarding point clouds in a logistic setting yet exists. Therefore, this paper investigates the generation of point clouds for logistics and the use of deep learning for classification and segmentation of the generated datasets. First, we generate different datasets regarding logistics with a virtual sensor. Afterwards, two state-of-the-art networks are evaluated and compared: PointNet++ and DGCN [2] [3]. Three different tasks are considered: The classification of logistic objects under the closed-world as well as the open-set assumption is assessed. Finally, the segmentation of logistic scenes is evaluated. Additionally, since simple surfaces can be removed reliably by traditional means, the effect of removing these as preprocessing step is evaluated as well. Both networks are able to reliably classify logistic objects with and without floor, even in the presence of unknown classes. However, while the segmentation performs well on average, some negative outliers do exist. Using transfer learning by pretraining the network with point clouds presenting the complete shape leads to a better performance.

This paper presents an effective two-step scheme for automatic object detection in computed radiography (CR) images. First, various structure elements of the morphological filters, designed by incorporating available morphological features of the objects of interest including their sizes and rough shape descriptions, are used to effectively distinguish the foreign object candidates from the complex background structures. Second, since the boundaries of the objects are the key features in reflecting object characteristics, active contour models are employed to accurately outline the morphological shapes of the suspicious foreign objects to further reduce the rate of false alarms. The actual detection scheme is accomplished by jointly using these two steps. The proposed methods are tested with a database of 50 hand-wrist computed radiographic images containing various types of foreign objects. Our experimental results demonstrate that the combined use of morphological filters and active contour models can provide an effective automatic detection of foreign objects in CR images achieving good sensitivity and specificity, and the accurate descriptions of the object morphological characteristics.

[abstFig src='/00280006/11.jpg' width='300' text='Road detection method with HOG and SVM' ] This paper describes a road area detection method using a support vector machine (SVM) and histogram of oriented gradient (HOG) features. The boundary lines have many features, such as changes in height, color, and brightness, but these are sensitive to noise. In terms of robustness, it is difficult to match road boundary lines with the boundary lines on 2D maps. Localization methods using texture matching are accurate, but they have disadvantages related to adapting to changes in the environment. We therefore decided to make a classifier to differentiate road areas from other areas by detecting the road plane. First, we calculate the HOG features from range data acquired by 3D LiDAR. We then create the road area classifier by applying SVM. Finally, we evaluate the basic performance of the proposed method in simulation and in the real world.