Frequency: Quarterly E- ISSN: 2277-6249 P- ISSN: Awaited Abstracted/ Indexed in: CNKI Scholar (China National Knowledge Infrastructure), Ulrichâ??s International Periodical Directory, Google Scholar, SCIRUS, getCITED, Genamics JournalSeek
Quarterly published â??Inventi Rapid/Impact: Machine Visionâ?? publishes high quality unpublished, as well as high impact pre-published research and reviews expanding to - algorithms, architectures, VLSI implementations, multidimensional and multisensor machine vision, real-time techniques, image databases, virtual reality and visualization. Its readership includes scientific, industrial, biomedical and military professionals.
One of the most important issues in human motion analysis is the tracking and 3D reconstruction of human motion, which\r\nutilizes the anatomic points� positions. These points can uniquely define the position and orientation of all anatomical segments.\r\nIn this work, a new method is proposed for tracking and 3D reconstruction of human motion from the image sequence of a\r\nmonocular static camera. In this method, 2D tracking is used for 3D reconstruction, which a database of selected frames is used\r\nfor the correction of tracking process. The method utilizes a new image descriptor based on discrete cosine transform (DCT),\r\nwhich is employed in different stages of the algorithm. The advantage of using this descriptor is the capabilities of selecting proper\r\nfrequency regions in various tasks, which results in an efficient tracking and pose matching algorithms. The tracking and matching\r\nalgorithms are based on reference descriptor matrixes (RDMs), which are updated after each stage based on the frequency regions\r\nin DCT blocks. Finally, 3D reconstruction is performed using Taylor�s method. Experimental results show the promise of the\r\nalgorithm....
Nowadays, research in autonomous underwater manipulation has demonstrated simple\napplications like picking an object from the sea floor, turning a valve or plugging and unplugging a\nconnector. These are fairly simple tasks compared with those already demonstrated by the mobile\nrobotics community, which include, among others, safe arm motion within areas populated with a\npriori unknown obstacles or the recognition and location of objects based on their 3D model to grasp\nthem. Kinect-like 3D sensors have contributed significantly to the advance of mobile manipulation\nproviding 3D sensing capabilities in real-time at low cost. Unfortunately, the underwater robotics\ncommunity is lacking a 3D sensor with similar capabilities to provide rich 3D information of the\nwork space. In this paper, we present a new underwater 3D laser scanner and demonstrate its\ncapabilities for underwater manipulation. In order to use this sensor in conjunction with manipulators,\na calibration method to find the relative position between the manipulator and the 3D laser scanner\nis presented. Then, two different advanced underwater manipulation tasks beyond the state of\nthe art are demonstrated using two different manipulation systems. First, an eight Degrees of\nFreedom (DoF) fixed-base manipulator system is used to demonstrate arm motion within a work\nspace populated with a priori unknown fixed obstacles. Next, an eight DoF free floating Underwater\nVehicle-Manipulator System (UVMS) is used to autonomously grasp an object from the bottom of a\nwater tank....
The manufacturing environment rapidly changes in turbulence fashion. Digital\r\nmanufacturing (DM) plays a significant role and one of the key strategies in setting up vision\r\nand strategic planning toward the knowledge based manufacturing. An approach of combining\r\n3D machine vision (3D-MV) and an Additive Manufacturing (AM) may finally be finding its\r\nniche in manufacturing. This paper briefly overviews the integration of the 3D machine vision\r\nand AM in concurrent product and process development, the challenges and opportunities, the\r\nimplementation of the 3D-MV and AM at POLMAN Bandung in accelerating product design\r\nand process development, and discusses a direct deployment of this approach on a real case\r\nfrom our industrial partners that have placed this as one of the very important and strategic\r\napproach in research as well as product/prototype development. The strategic aspects and\r\nneeds of this combination approach in research, design and development are main concerns of\r\nthe presentation....
Technological solutions for obstacle-detection systems have been proposed to prevent accidents in safety-transport applications. In order to avoid the limits of these proposed technologies, an obstacle-detection system utilizing stereo cameras is proposed to detect and localize multiple objects at level crossings. Background subtraction is first performed using the color independent component analysis technique, which has proved its performance against other well-known object-detection methods. The main contribution is the development of a robust stereo-matching algorithm which reliably localizes in 3D each segmented object. A standard stereo dataset and real-world images are used to test and evaluate the performances of the proposed algorithm to prove the efficiency and the robustness of the proposed video-surveillance system....
Urban traffic congestion is a severe and widely studied problem over the\ndecade because of the negative impacts. However, in recent years some approaches\nemerge as proper and suitable solutions. The Carpooling initiative is\none of the most representative efforts to propitiate a responsible use of particular\nvehicles. Thus, the paper introduces a carpooling model considering\nthe usersâ?? preference to reach an appropriate match among drivers and passengers.\nIn particular, the paper conducts a study of 6 of the most avid classified\ntechniques in machine learning to create a model for the selection of travel\ncompanions. The experimental results show the modelsâ?? precision and assess\nthe best cases using Friedmanâ??s test. Finally, the conclusions emphasize\nthe relevance of the proposed study and suggest that it is necessary to extend\nthe proposal with more drives and passengersâ?? data....
Image classification is an important problem in computer vision. The sparse coding spatial\npyramid matching (ScSPM) framework is widely used in this field. However, the sparse coding\ncannot effectively handle very large training sets because of its high computational complexity,\nand ignoring the mutual dependence among local features results in highly variable sparse codes\neven for similar features. To overcome the shortcomings of previous sparse coding algorithm,\nwe present an image classification method, which replaces the sparse dictionary with a stable\ndictionary learned via low computational complexity clustering, more specifically, a k-medoids\ncluster method optimized by k-means++. The proposed method can reduce the learning complexity\nand improve the featureâ??s stability. In the experiments, we compared the effectiveness of our method\nwith the existing ScSPM method and its improved versions. We evaluated our approach on two\ndiverse datasets: Caltech-101 and UIUC-Sports. The results show that our method can increase the\naccuracy of spatial pyramid matching, which suggests that our method is capable of improving\nperformance of sparse coding features....
With explosive growth of malware, Internet users face enormous threats from Cyberspace, known as â??fifth dimensional space.â?\nMeanwhile, the continuous sophisticatedmetamorphismofmalware such as polymorphismand obfuscationmakes itmore difficult\nto detect malicious behavior. In the paper, based on the dynamic feature analysis of malware, a novel feature extraction method\nof hybrid gram (H-gram) with cross entropy of continuous overlapping subsequences is proposed, which implements semantic\nsegmentation of a sequence of API calls or instructions. The experimental results show the H-gram method can distinguish\nmalicious behaviors and is more effective than the fixed-length n-gram in all four performance indexes of the classification\nalgorithms such as ID3, Random Forest, AdboostM1, and Bagging....
This work demonstrates how a high throughput\nrobotic machine vision systems can quantify seedling\ndevelopment with high spatial and temporal resolution.The\nthroughput that the system provides is high enough to match\nthe needs of functional genomics research. Analyzing images\nof plant seedlings growing and responding to stimuli is a\nproven approach to finding the effects of an affected gene.\nHowever, with 104 genes in a typical plant genome, comprehensive\nstudies will require high throughput methodologies.\nTo increase throughput without sacrificing spatial or\ntemporal resolution, a 3 axis robotic gantry system utilizing\nvisual servoing was developed. The gantry consists of\ndirect drive linear servo motors that can move the cameras\nat a speed of 1 m/s with an accuracy of 1 ?m, and a repeatability\nof 0.1 ?m. Perpendicular to the optical axis of the\ncameras was a 1 m2 sample fixture holds 36 Petri plates in\nwhich 144 Arabidopsis thaliana seedlings (4 per Petri plate)\ngrew vertically along the surface of an agar gel. A probabilistic\nimage analysis algorithm was used to locate the root\nof seedlings and a normalized gray scale variance measure\nwas used to achieve focus by servoing along the optical axis.\nRotation of the sample holder induced a gravitropic bending\nresponse in the roots, which are approximately 45 ?m wide\nand several millimeter in length. The custom hardware and\nsoftware described here accurately quantified the gravitropic\nresponses of the seedlings in parallel at approximately 3 min\nintervals over an 8-h period. Here we present an overview of\nour system and describe some of the necessary capabilities\nand challenges to automating plant phenotype studies....
State-of-the-art human detection methods focus on deep network architectures to achieve\nhigher recognition performance, at the expense of huge computation. However, computational\nefficiency and real-time performance are also important evaluation indicators. This paper presents\na fast real-time human detection and flow estimation method using depth images captured by a\ntop-view TOF camera. The proposed algorithm mainly consists of head detection based on local\npooling and searching, classification refinement based on human morphological features, and tracking\nassignment filter based on dynamic multi-dimensional feature. A depth image dataset record\nwith more than 10k entries and departure events with detailed human location annotations is\nestablished. Taking full advantage of the distance information implied in the depth image, we achieve\nhigh-accuracy human detection and people counting with accuracy of 97.73% and significantly reduce\nthe running time. Experiments demonstrate that our algorithm can run at 23.10 ms per frame on a\nCPU platform. In addition, the proposed robust approach is effective in complex situations such as\nfast walking, occlusion, crowded scenes, etc....
Personalized or precision medicine is a new paradigm that holds great promise for\nindividualized patient diagnosis, treatment, and care. However, personalized medicine has only\nbeen described on an informal level rather than through rigorous practical guidelines and statistical\nprotocols that would allow its robust practical realization for implementation in day-to-day clinical\npractice. In this paper, we discuss three key factors, which we consider dimensions that effect\nthe experimental design for personalized medicine: (I) phenotype categories; (II) population size;\nand (III) statistical analysis. This formalization allows us to define personalized medicine from a\nmachine learning perspective, as an automized, comprehensive knowledge base with an ontology\nthat performs pattern recognition of patient profiles....
Loading....