Showing papers on "Monocular vision published in 2012"

Mobile robot monocular vision navigation based on road region and boundary estimation

Abstract: We present a monocular vision-based navigation system that incorporates two contrasting approaches: region segmentation that computes the road appearance, and road boundary detection that estimates the road shape. The former approach segments the image into multiple regions, then selects and tracks the most likely road appearance. On the other hand, the latter detects the vanishing point and road boundaries to estimate the shape of the road. Our algorithm operates in urban road settings and requires no training or camera calibration to maximize its adaptability to many environments. We tested our system in 1 indoor and 3 outdoor urban environments using our ground-based robot, Beobot 2.0, for real-time autonomous visual navigation. In 20 trial runs the robot was able to travel autonomously for 98.19% of the total route length of 316.60m.

On-board lane detection system for intelligent vehicle based on monocular vision

Abstract: The objective of this research is to develop a monocular vision system that can locate the positions of the road lane in real time. First, Canny approach is used to obtain edge map from the road image acquired from monocular camera mount on vehicle; Second, a matching process is conducted to normalize the candidates of road line; Third, a searching method is used for reinforce potential road lines while degraded those impossible ones; Forth, a linking condition is used to further enhance the confidence of the potential lane lines, and a K-means cluster algorithm is employed to localize the lane lines; Finally, a on board system is designed for experiment. The proposed system is shown to work well under various conditions on the roadway. Besides, the computation cost is inexpensive and the system's response is almost real time.

Road direction detection based on vanishing-point tracking

Abstract: We present a novel approach for vision-based road direction detection for autonomous Unmanned Ground Vehicles (UGVs). The proposed method utilizes only monocular vision information similar to human perception to detect road directions with respect to the vehicle. The algorithm searches for a global feature of the roads due to perspective projection (so-called vanishing point) to distinguish road directions. The proposed approach consists of two stages. The first stage estimates the vanishing-point locations from single frames. The second stage uses a Rao-Blackwellised particle filter to track initial vanishing-point estimations over a sequence of images in order to provide more robust estimation. Simultaneously, the direction of the road ahead of the vehicle is predicted, which is prerequisite information for vehicle steering and path planning. The proposed approach assumes minimum prior knowledge about the environment and can cope with complex situations such as ground cover variations, different illuminations, and cast shadows. Its performance is evaluated on video sequences taken during test run of the DARPA Grand Challenge.

Method for measuring and pre-warning lane departure distance based on monocular vision

Abstract: The invention discloses a method for measuring and pre-warning a lane departure distance based on monocular vision, belonging to the technical field of computer imaging processing. The method comprises the following steps of: collecting a video image through a monocular video camera installed in the front of an automobile at first, completing the detection of a lane line after processing through an image processing technology, and extracting geometrical information of the lane line; obtaining vertical distances between the automobile and the lane lines at left and right sides by utilizing thethree-dimensional geometry transformation relation of a pinhole imaging principle; and establishing a departure pre-warning decision method according to the vertical distances measured in real time, and providing effective information for an intelligent assistant driving technology. According to the method disclosed by the invention, when the lane line is detected by utilizing Hough transform, a constraint condition is added, a part of virtual lane line is excluded, and the operation speed and the lane line detection accuracy are increased; simultaneously, the lane departure pre-warning can be realized only by utilizing image information; the measurement influence of a vehicle departure angle on the lane departure distance is low; furthermore, the solving operation speed is high owing to the use of the three-dimensional geometry transform method; and the requirements of the intelligent assistant driving technology can be satisfied.

Image appearance based loop closure detecting method in monocular vision SLAM (simultaneous localization and mapping)

Abstract: The invention discloses an image appearance based loop closure detecting method in monocular vision SLAM (simultaneous localization and mapping). The image appearance based loop closure detecting method includes acquiring images of the current scene by a monocular camera carried by a mobile robot during advancing, and extracting characteristics of bag of visual words of the images of the current scene; preprocessing the images by details of measuring similarities of the images according to inner products of image weight vectors and rejecting the current image highly similar to a previous history image; updating posterior probability in a loop closure hypothetical state by a Bayesian filter process to carry out loop closure detection so as to judge whether the current image is subjected to loop closure or not; and verifying loop closure detection results obtained in the previous step by an image reverse retrieval process. Further, in a process of establishing a visual dictionary, the quantity of clustering categories is regulated dynamically according to TSC (tightness and separation criterion) values which serve as an evaluation criterion for clustering results. Compared with the prior art, the loop closure detecting method has the advantages of high instantaneity and detection precision.

Monocular vision and IMU based navigation for a small unmanned helicopter

Abstract: Focusing on the problem of autonomous flight for a small unmanned helicopter in GPS-denied environments, this paper proposes a navigation system which combines monocular visual and inertial measurements. The homography based visual odometry, as well as the inertial measurement from an IMU, can estimate the motion of a small unmanned helicopter. An extended Kalman filter is used to fuse the estimations of both visual and inertial sensors to provide the available navigation data. Finally, the effectiveness of this system is verified by real experiments. The Vision/IMU navigation has almost the same performance as the traditional GPS/IMU navigation.

Improving robustness of robotic grasping by fusing multi-sensor

Abstract: Since the visual system is susceptible to the lighting condition and surroundings changes, the accuracy for object localization of robot grasping system based on visual servo is rather poor so as to the low grasping success rate and bad robustness of the whole system In view of such phenomenon, in this paper, we propose a method of fusing binocular camera accompany with monocular vision, IR sensors, tactile sensors and encoders to design a reliable and robust grasping system that could offer real-time feedback information In order to avoid the situation of robot grasping-nothing, we use the binocular vision supplemented by monocular camera and IR sensors to locate accurately By analyzing the contact model and pressure between gripper and the object, a durable, non-slip rubber coating is designed to increase the fingertip's friction, What's more, Fuzzy Neural Network (FNN) method was applied to fuse the information of multiple sensors in our robot system By monitoring force and position information in the process of grasping all the time, the system can reduce the phenomenon of slippage and crush of object as well as improve the grasping stability greatly The experimental results show the effectiveness of our system

Monocular vision based 6D object localization for service robot's intelligent grasping

Abstract: Intelligent grasping is still a hard problem for home service robots. There are two major issues in the intelligent grasping, i.e. the object recognition and the pose estimation. To grasp casually placed objects, the robot needs the object's full 6 degrees of freedom pose data. To deal with the challenges such as illumination changes, cluttered background, occlusion, etc., we propose a monocular vision based object recognition and 6D pose estimation method. The SIFT feature point matching and brute-force search algorithm is used to do a tentative object recognition. The object recognition result is then verified with the homography constraint. After passing the verification, the 6D pose estimation is obtained through the decomposition of the homography matrix and the result is refined using the Levenberg-Marquardt algorithm. We embed our pose estimation method in a tracking by detection framework to keep computing and refining the pose during the whole approaching procedure. To test our method, a robot arm of seven degrees of freedom was utilized for a group of grasping experiments. The experimental results showed that our approach successfully recognized and grasped a variety of household objects with decent accuracy.

Building a 3D scanner system based on monocular vision.

Abstract: This paper proposes a three-dimensional scanner system, which is built by using an ingenious geometric construction method based on monocular vision. The system is simple, low cost, and easy to use, and the measurement results are very precise. To build it, one web camera, one handheld linear laser, and one background calibration board are required. The experimental results show that the system is robust and effective, and the scanning precision can be satisfied for normal users.

Monocular SLAM for Visual Odometry: A Full Approach to the Delayed Inverse-Depth Feature Initialization Method

Abstract: This paper describes in a detailed manner a method to implement a simultaneous localization and mapping (SLAM) system based on monocular vision for applications of visual odometry, appearance-based sensing, and emulation of range-bearing measurements. SLAM techniques are required to operate mobile robots in a priori unknown environments using only on-board sensors to simultaneously build a map of their surroundings; this map will be needed for the robot to track its position. In this context, the 6-DOF (degree of freedom) monocular camera case (monocular SLAM) possibly represents the harder variant of SLAM. In monocular SLAM, a single camera, which is freely moving through its environment, represents the sole sensory input to the system. The method proposed in this paper is based on a technique called delayed inverse-depth feature initialization, which is intended to initialize new visual features on the system. In this work, detailed formulation, extended discussions, and experiments with real data are presented in order to validate and to show the performance of the proposal.

Importance of binocular vision in foot placement accuracy when stepping onto a floor-based target during gait initiation

Abstract: This study investigated the importance of binocular vision to foot placement accuracy when stepping onto a floor-based target during gait initiation. Starting from stationary, participants placed alternate feet onto targets sequentially positioned along a straight travel path with the added constraint that the initial target (target 1) could move in the medio-lateral (M-L) direction. Repeated trials when target 1 remained stationary or moved laterally at the instant of lead-limb toe-off (TO) or 200 ms after TO (early swing) were undertaken under binocular and monocular viewing. Catch trials when target 1 shifted medially were also undertaken. Foot-reach kinematics, foot trajectory corrections and foot placement accuracy for the step onto target 1 were determined via 3D motion analyses. Peak foot-reach velocity and initial foot-reach duration were unaffected by vision condition but terminal foot-reach duration was prolonged under monocular conditions (p = 0.002). Foot trajectory alteration onsets were unaffected by vision condition, but onsets occurred sooner when the target shifted in early swing compared to at TO (p = 0.033). M-L foot placement accuracy decreased (p = 0.025) and variability increased (p = 0.05) under monocular conditions, particularly when stepping onto the moving target. There was no difference between vision conditions in A-P foot placement accuracy. Results indicate that monocular vision provides sufficient information to determine stepping distance and correctly transport the foot towards the target but binocular vision is required to attain a precise M-L foot placement; particularly so when stepping onto a moving target. These findings are in agreement with those found in the reaching and grasping literature, indicating that binocular vision is important for end-point precision.

Robot Navigation Control Based on Monocular Images: An Image Processing Algorithm for Obstacle Avoidance Decisions

Abstract: This paper covers the use of monocular vision to control autonomous navigation for a robot in a dynamically changing environment. The solution focused on using colour segmentation against a selected floor plane to distinctly separate obstacles from traversable space: this is then supplemented with canny edge detection to separate similarly coloured boundaries to the floor plane. The resulting binary map (where white identifies an obstacle-free area and black identifies an obstacle) could then be processed by fuzzy logic or neural networks to control the robot’s next movements. Findings show that the algorithm performed strongly on solid coloured carpets, wooden, and concrete floors but had difficulty in separating colours in multicoloured floor types such as patterned carpets.

Detection of moving objects in image plane for robot navigation using monocular vision

Abstract: This article presents an algorithm for moving object detection (MOD) in robot visual simultaneous localization and mapping (SLAM). This MOD algorithm is designed based on the defining epipolar constraint for the corresponding feature points on image plane. An essential matrix obtained using the state estimator is utilized to represent the epipolar constraint. Meanwhile, the method of speeded-up robust feature (SURF) is employed in the algorithm to provide a robust detection for image features as well as a better description of landmarks and of moving objects in visual SLAM system. Experiments are carried out on a hand-held monocular camera to verify the performances of the proposed algorithm. The results show that the integration of MOD and SURF is efficient for robot navigating in dynamic environments.

Multiple episodes of convergence in genes of the dim light vision pathway in bats.

Abstract: The molecular basis of the evolution of phenotypic characters is very complex and is poorly understood with few examples documenting the roles of multiple genes. Considering that a single gene cannot fully explain the convergence of phenotypic characters, we choose to study the convergent evolution of rod vision in two divergent bats from a network perspective. The Old World fruit bats (Pteropodidae) are non-echolocating and have binocular vision, whereas the sheath-tailed bats (Emballonuridae) are echolocating and have monocular vision; however, they both have relatively large eyes and rely more on rod vision to find food and navigate in the night. We found that the genes CRX, which plays an essential role in the differentiation of photoreceptor cells, SAG, which is involved in the desensitization of the photoactivated transduction cascade, and the photoreceptor gene RH, which is directly responsible for the perception of dim light, have undergone parallel sequence evolution in two divergent lineages of bats with larger eyes (Pteropodidae and Emballonuroidea). The multiple convergent events in the network of genes essential for rod vision is a rare phenomenon that illustrates the importance of investigating pathways and networks in the evolution of the molecular basis of phenotypic convergence.

Cataract surgery: emotional reactions of patients with monocular versus binocular vision

Abstract: PURPOSE: To analyze emotional reactions related to cataract surgery in two groups of patients (monocular vision - Group 1; binocular vision - Group 2). METHODS: A transversal comparative study was performed using a structured questionnaire from a previous exploratory study before cataract surgery. RESULTS: 206 patients were enrolled in the study, 96 individuals in Group 1 (69.3 ± 10.4 years) and 110 in Group 2 (68.2 ± 10.2 years). Most patients in group 1 (40.6%) and 22.7% of group 2, reported fear of surgery (p<0.001). The most important causes of fear were: possibility of blindness, ocular complications and death during surgery. The most prevalent feelings among the groups were doubts about good results and nervousness. CONCLUSION: Patients with monocular vision reported more fear and doubts related to surgical outcomes. Thus, it is necessary that phisycians considers such emotional reactions and invest more time than usual explaining the risks and the benefits of cataract surgery.Ouvir

Tumbling target reconstruction and pose estimation through fusion of monocular vision and sparse-pattern range data

Abstract: A framework for 3D target reconstruction and relative pose estimation through fusion of vision and sparse-pattern range data (eg line-scanning LIDAR) is presented The algorithm augments previous work in monocular vision-only SLAM/SfM to incorporate range data into the overall solution The aim of this work is to enable a more dense reconstruction with accurate relative pose estimation that is unambiguous in scale In order to incorporate range data, a linear estimator is presented to estimate the overall scale factor using vision-range correspondence A motivating mission is the use of resource-constrained micro- and nano-satellites to perform autonomous rendezvous and docking operations with uncommunicative, tumbling targets, about which little or no prior information is available The rationale for the approach is explained, and an algorithm is presented The implementation using a modified Rao-Blackwellised particle filter is described and tested Results from numerical simulations are presented that demonstrate the performance and viability of the approach

Manual convoying of automated urban vehicles relying on monocular vision

Abstract: This paper deals with platooning navigation in the context of innovative solutions for urban transportation systems More precisely, the case of a manually driven vehicle leading a convoy of automated ones is considered Vehicle localization relies solely on monocular vision: a 3D map of the environment is built beforehand from reference video sequences, and then used to derive vehicle absolute location from the current camera image The 3D vision map presents however distortions wrt a metric world, but these latter can be shown to be locally homogeneous They can then be accurately corrected via a 1-dim function evaluated with a nonlinear observer relying on odometric data Next, the platoon reference trajectory is built as a B-Spline curve extended on-line via local optimization from the successive locations of the lead vehicle, and a global decentralized control strategy, supported by intervehicle communication, is designed to achieve accurate platooning with no oscillation within the convoy Experimental results, carried out with two urban vehicles, demonstrate the capabilities of the proposed approach

Circle object recognition based on monocular vision for home security robot

Abstract: In this paper, a circle object recognition method based on monocular vision for the home security robots is proposed. This vision system is able to process image and recognize a colored ball rapidly. The proposed method consists of two submodules, which are the object segmentation module and the circle detection module. In order to implement the object segmentation, the color feature is applied to find out the region of the object. After the region of the object is determined, a fast randomized circle detection method is used by checks if there have enough number radius which are the same in a circle in the region. Because of the double recognition, this system can improve the precision for detecting a colored ball. The proposed method is tested on a home security robot to find a ball. The experimental results illustrate the effectiveness of the proposed method.

Seamless aiding of inertial-slam using Visual Directional Constraints from a monocular vision

Abstract: Inertial-SLAM has been actively studied as it can provide all-terrain navigational capability with full six degrees-of-freedom information to autonomous robots. With the recent availability of low-cost inertial and vision sensors, a light-weight and accurate mapping system could be achieved for many robotic tasks such as land/aerial explorations. The key challenge toward this is in the availability of reliable and constant aiding information to correct the inertial system which is intrinsically unstable. The existing approaches have been relying on feature-based maps, which require accurate depth-resolution process to correct the inertial units properly where the aiding rate is highly dependent on the map density. In this work we propose to directly integrate the visual odometry to the inertial system by fusing the scale ambiguous translation vectors as Visual Directional Constraints (VDC) on vehicle motion at high update rates, while the 3D map being still used to constrain the longitudinal drifts but in a relaxed way. In this way, the visual odometry information can be seamlessly fused to inertial system by resolving the scale ambiguity problem between inertial and monocular camera thus achieving a reliable and constant aiding. The proposed approach is evaluated on SLAM benchmark dataset and simulated environment, showing a more stable and consistent performance of monocular inertial-SLAM.

Real-time on-road vehicle detection algorithm based on monocular vision

Abstract: Increasing driving safety by virtue of advanced technology requires real-time and accurate detection of vehicles in far and close distance, which pose a threat to the host vehicle This paper presents a real-time on-road vehicle detection algorithm based on monocular vision First, auto-adapted threshold segmentation is proposed to extract shadow features Then, a special mask is used for morphology computing to retain the features of vehicles in far and close distance And, vanishing point constraint is applied for the fast verification of vehicles Finally, the tracking of vehicles assists to stabilize the detection results The experiments show that the average processing speed reaches 39 frames/ms, and the fast detection of vehicle under different weather conditions in the day time can also work accurately

Real-time performance of a hands-free semi-autonomous wheelchair system using a combination of stereoscopic and spherical vision

Abstract: This paper is concerned with the operational performance of a semi-autonomous wheelchair system named TIM (Thought-controlled Intelligent Machine), which uses cameras in a system configuration modeled on the vision system of a horse. This new camera configuration utilizes stereoscopic vision for 3-Dimensional (3D) depth perception and mapping ahead of the wheelchair, combined with a spherical camera system for 360-degrees of monocular vision. The unique combination allows for static components of an unknown environment to be mapped and any surrounding dynamic obstacles to be detected, during real-time autonomous navigation, minimizing blind-spots and preventing accidental collisions with people or obstacles. Combining this vision system with a shared control strategy provides intelligent assistive guidance during wheelchair navigation, and can accompany any hands-free wheelchair control technology for people with severe physical disability. Testing of this system in crowded dynamic environments has displayed the feasibility and real-time performance of this system when assisting hands-free control technologies, in this case being a proof-of-concept brain-computer interface (BCI).

An Optimization Based Framework for Human Pose Estimation in Monocular Videos

Abstract: Human pose estimation using monocular vision is a challenging problem in computer vision. Past work has focused on developing efficient inference algorithms and probabilistic prior models based on captured kinematic/dynamic measurements. However, such algorithms face challenges in generalization beyond the learned dataset.

method for rapidly measuring volume of object

Abstract: The invention provides a method for measuring the volume of an object by using a monocular vision positioning principle and a two-dimensional laser scanner. According to the technical scheme provided by the invention, a monocular vision positioning system and a two-dimensional laser scanner frame are arranged at any side of a wheel type vehicle and the wheel type vehicle travels around an object for a circle. The method comprises the following steps of: step 1, collecting data; step 2, calculating a position coordinate of a projective point of the two-dimensional laser scanner on a road surface at each positioning time; step 3, calculating the position coordinate of a projective point of the two-dimensional laser scanner on the road surface when each object outline is obtained; and step 4, obtaining a three-dimensional reconstructed result of the object and calculating the volume of the object. The method provided by the invention has the beneficial effects that: omnibearing rapid scanning can be carried out on the objects of irregular shapes, the volume of the object can be obtained rapidly and accurately, and the automation degree is high. Moreover, according to the invention, the calculation method is simpler and equipment is convenient to use.

Monocular vision single image positioning method based on parallelogram

Abstract: The invention discloses a monocular vision single image positioning method based on a parallelogram. An object is spatially positioned through four characteristic points with parallelogram geometric constraint on a single image only. The monocular vision single image positioning method comprises the following steps of: (1) acquiring a parallelogram image through a visual sensor, and acquiring coordinates of imaging points of four vertexes of the parallelogram in a camera coordinate system through an image processing algorithm; (2) obtaining a normal of a plane where the parallelogram is positioned in the camera coordinate system by means of two vanishing points; (3) acquiring coordinates of the four vertexes of the parallelogram in the camera coordinate system according to a transparent imaging model on the basis of the step (2); and (4) acquiring a spatial position of the object according to the coordinates of the four vertexes of the parallelogram in the camera coordinate system and coordinates of the four vertexes of the parallelogram in an object coordinate system. The space object is positioned by using a camera, and the economical efficiency is high; and moreover, when the position of the space object is calculated, the calculation process is greatly simplified by utilizing the characteristics of two vanishing points of the parallelogram imaging.