Binocular Vision Technology Research Articles

An obstacle avoidance safety detection algorithm for power lines combining binocular vision technology and improved object detection

In this paper, a framework of obstacle avoidance algorithm applied to power line damage safety distance detection is constructed, and its overall architecture and key processes are described in detail. The system design covers three core modules: visual data acquisition and preliminary processing, accurate target recognition and distance measurement, and system error analysis and correction. In the visual data processing chain, we deeply analyze every step from image acquisition to preprocessing to feature extraction, aiming to enhance the adaptability of applications to complex scenes. The target recognition and distance estimation part integrates advanced technology of deep learning to improve the reliability of recognition accuracy and distance estimation. In addition, many common error sources, such as system bias, parallax discontinuity, fluctuation of illumination conditions, etc., are discussed in depth, and corresponding correction strategies are proposed to ensure the accuracy and stability of the system, which provides powerful technical support for achieving efficient and accurate safety monitoring. Specifically, by carefully adjusting the learning rate, convolution kernel size, batch size, pooling layer type, and number of hidden layer nodes, we succeeded in improving the overall accuracy from the initial average of 92.4–95%, and the error rate decreased accordingly.

Omni-directional attitude detection of advanced hydraulic support relative to roadway based on visual measurement principle

In order to address the challenge of detecting the attitude of advanced hydraulic support in a harsh mine environment, this study proposes a binocular vision technology that accurately measures the omni-directional attitude of advanced hydraulic support relative to the roadway in real-time. The first step involves constructing a visual measurement system and optimizing the software development platform based on the actual motion conditions. Next, an innovative algorithm for space point reconstruction and omni-directional attitude is proposed. Finally, an indoor ground experimental platform is built to demonstrate the feasibility and effectiveness of the method by comparing it with high-precision experiments. The results show that the average absolute deviation of the measured attitude angle is within 0.4° and the average absolute deviation of the measured distance is within 6 mm, which meets the accuracy requirements of the fully mechanized mining equipment under the mine. This method has high measurement accuracy and sufficient visibility, which has important practical significance for ensuring the support efficiency of advanced hydraulic support and the intelligent production of working faces.

Design and Experiment of an Unoccupied Control System for a Tracked Grain Vehicle.

In order to enhance crop harvesting efficiency, an automatic-driving tracked grain vehicle system was designed. Based on the harvester chassis, we designed the mechanical structure of a tracked grain vehicle with a loading capacity of 4.5 m3 and a grain unloading hydraulic system. Using the BODAS hydraulic controller, we implemented the design of an electronic control system that combines the manual and automatic operation of the chassis walking mechanism and grain unloading mechanism. We utilized a hybrid A* algorithm to plan the traveling path of the tracked grain vehicle, and the path-tracking controller of the tracked grain vehicle was designed by combining fuzzy control and pure pursuit algorithms. Leveraging binocular vision technology and semantic segmentation technology, we designed an automatic grain unloading control system with functions of grain tank recognition and grain unloading regulation control. Finally, we conducted experiments on automatic grain unloading control and automatic navigation control in the field. The results showed that both the precision of the path-tracking control and the automatic unloading system meet the requirements for practical unoccupied operations of the tracked grain vehicle.

Real-time binocular visual localization system based on the improved BGNet stereo matching framework.

Binocular vision technology is widely used to acquire three-dimensional information of images because of its low cost. In recent years, the use of deep learning for stereo matching has shown promising results in improving the measurement stability of binocular vision systems, but the real-time performance in high-precision networks is typically poor. Therefore, this study constructed a deep-learning-based stereo matching binocular vision system based on the BGLGA-Net, which combines the advantages of past networks. Experiments showed that the ability to detect the edges of foreground objects was enhanced. The network was used to build a system on the Xavier NX. The measurement accuracy and stability were better than those of traditional algorithms.

Headland Identification and Ranging Method for Autonomous Agricultural Machines

Headland boundary identification and ranging are the key supporting technologies for the automatic driving of intelligent agricultural machinery, and they are also the basis for controlling operational behaviors such as autonomous turning and machine lifting. The complex, unstructured environments of farmland headlands render traditional image feature extraction methods less accurate and adaptable. This study utilizes deep learning and binocular vision technologies to develop a headland boundary identification and ranging system built upon the existing automatic guided tractor test platform. A headland image annotation dataset was constructed, and the MobileNetV3 network, notable for its compact model structure, was employed to achieve binary classification recognition of farmland and headland images. An improved MV3-DeeplabV3+ image segmentation network model, leveraging an attention mechanism, was constructed, achieving a high mean intersection over union (MIoU) value of 92.08% and enabling fast and accurate detection of headland boundaries. Following the detection of headland boundaries, binocular stereo vision technology was employed to measure the boundary distances. Field experiment results indicate that the system’s average relative errors of distance in ranging at distances of 25 m, 20 m, and 15 m are 6.72%, 4.80%, and 4.35%, respectively. This system is capable of meeting the real-time detection requirements for headland boundaries.

A coaxial multi-ring detection method for measuring the pitch and thickness accuracy of cylindrical gears

Performance of gear transmissions affects the performance of mechanical equipment. It is necessary to develop more reliable gear pitch and tooth thickness accuracy detection methods in order to evaluate gear transmission performance and detect gear pitch and tooth thickness accuracy more accurately. Based on the basic theory of gears, binocular vision technology, and statistical principles, a new method that measures gear pitch and tooth thickness using machine vision is proposed: coaxial multi-ring detection (MCD). There is no contact, no damage, and a high degree of efficiency with this method. Using this method, we are able to detect the pitch and thickness accuracy of each gear tooth multiple times within the tooth width range and in multiple directions perpendicular to the gear axis. We statistically analyze the measurement results to determine the gear’s most accurate detection results. The measurement method for gear machining accuracy is investigated using the coaxial multi-ring detection method. The statistical analysis of multiple measurement results is carried out, and the measurement results obtained are highly consistent with those of the gear detection center. In conclusion, the measurement results of this method are highly reliable, and they can be used as a reliable basis for evaluating gear transmission performance.

Research on Key Technologies of Smart Breeding Platform Based on 5G

Abstract In modern animal husbandry, the application of 5G technology opens a new era of farming management, especially in intelligent farming. The high-speed and low-latency communication characteristics make remote monitoring, data transmission, and thoughtful decision-making possible, significantly improving breeding efficiency and management. It has great potential for monitoring animal health, preventing and controlling diseases, and improving production efficiency. This paper explores the application of 5G technology in an intelligent farming management platform and its key technologies. By analyzing the needs of smart farming, we clarify the key application areas of 5G technology. To identify sick pigs, measure their weight accurately using binocular vision technology, and combine with machine learning algorithms, the YOLO v3 algorithm is utilized. The accuracy of sick pig identification based on YOLO v3 reaches 98%, the error of pig weight measurement is controlled within ±2%, and the real-time data transmission and processing realized by 5G technology significantly improves the efficiency of disease prevention and control.5G technology can effectively support the efficient operation of the intelligent farming platform, which is of great significance to enhance the level of farming management and promote the sustainable development of the animal husbandry industry.

Real-Time Counting and Height Measurement of Nursery Seedlings Based on Ghostnet–YoloV4 Network and Binocular Vision Technology

Traditional nursery seedling detection often uses manual sampling counting and height measurement with rulers. This is not only inefficient and inaccurate, but it requires many human resources for nurseries that need to monitor the growth of saplings, making it difficult to meet the fast and efficient management requirements of modern forestry. To solve this problem, this paper proposes a real-time seedling detection framework based on an improved YoloV4 network and binocular camera, which can provide real-time measurements of the height and number of saplings in a nursery quickly and efficiently. The methodology is as follows: (i) creating a training dataset using a binocular camera field photography and data augmentation; (ii) replacing the backbone network of YoloV4 with Ghostnet and replacing the normal convolutional blocks of PANet in YoloV4 with depth-separable convolutional blocks, which will allow the Ghostnet–YoloV4 improved network to maintain efficient feature extraction while massively reducing the number of operations for real-time counting; (iii) integrating binocular vision technology into neural network detection to perform the real-time height measurement of saplings; and (iv) making corresponding parameter and equipment adjustments based on the specific morphology of the various saplings, and adding comparative experiments to enhance generalisability. The results of the field testing of nursery saplings show that the method is effective in overcoming noise in a large field environment, meeting the load-carrying capacity of embedded mobile devices with low-configuration management systems in real time and achieving over 92% accuracy in both counts and measurements. The results of these studies can provide technical support for the precise cultivation of nursery saplings.

Non-contact measurement method for reconstructing three-dimensional scour depth field based on binocular vision technology in laboratory

The scour of pile foundations significantly impacts the bridge structure safety. This study proposes a three-dimensional (3D) non-contact scour pit depth measurement method. Cameras were calibrated using a three-step calibration procedure based on the diagonal length. Refraction errors caused by the interface between water and air were corrected. The corresponding solutions are given to the problem that it is difficult to mark sand with traditional speckles, and the measurement range of a single camera is limited. The matching of virtual grid points between different views at different moments was achieved using the theory of digital image correlation (DIC) based on zero-mean normalized cross-correlation (ZNCC). The scour pit was reconstructed by splicing the depth field from different views. The comparison between the results of direct and visual measurements at a single point proves that the difference between them was basically at the millimeter level. Scour pits of different water depths, velocities, reciprocating flows, and unidirectional flows were photographed and reconstructed. Based on these results, the development of scour pits under various flow conditions is discussed.

Deep-Learning-Based Wireless Visual Sensor System for Shiitake Mushroom Sorting.

The shiitake mushroom is the second-largest edible mushroom in the world, with a high nutritional and medicinal value. The surface texture of shiitake mushrooms can be quite different due to different growing environments, consequently leading to fluctuating market prices. To maximize the economic profit of the mushroom industry, it is necessary to sort the harvested mushrooms according to their qualities. This paper aimed to develop a deep-learning-based wireless visual sensor system for shiitake mushroom sorting, in which the visual detection was realized by the collection of images and cooperative transmission with the help of visual sensors and Wi-Fi modules, respectively. The model training process was achieved using Vision Transformer, then three data-augmentation methods, which were Random Erasing, RandAugment, and Label Smoothing, were applied under the premise of a small sample dataset. The training result of the final model turned out nearly perfect, with an accuracy rate reaching 99.2%. Meanwhile, the actual mushroom-sorting work using the developed system obtained an accuracy of 98.53%, with an 8.7 ms processing time for every single image. The results showed that the system could efficiently complete the sorting of shiitake mushrooms with a stable and high accuracy. In addition, the system could be extended for other sorting tasks based on visual features. It is also possible to combine binocular vision and multisensor technology with the current system to deal with sorting work that requires a higher accuracy and minor feature identification.

Trajectory Tracking and Load Monitoring for Moving Vehicles on Bridge Based on Axle Position and Dual Camera Vision

Monitoring traffic loads is vital for ensuring bridge safety and overload controlling. Bridge weigh-in-motion (BWIM) technology, which uses an instrumented bridge as a scale platform, has been proven as an efficient and durable vehicle weight identification method. However, there are still challenges with traditional BWIM methods in solving the inverse problem under certain circumstances, such as vehicles running at a non-constant speed, or multiple vehicle presence. For conventional BWIM systems, the velocity of a moving vehicle is usually assumed to be constant. Thus, the positions of loads, which are vital in the identification process, is predicted from the acquired speed and axle spacing by utilizing dedicated axle detectors (installed on the bridge surface or under the bridge soffit). In reality, vehicles may change speed. It is therefore difficult or even impossible for axle detectors to accurately monitor the true position of a moving vehicle. If this happens, the axle loads and bridge response cannot be properly matched, and remarkable errors can be induced to the influence line calibration process and the axle weight identification results. To overcome this problem, a new BWIM method was proposed in this study. This approach estimated the bridge influence line and axle weight by associating the bridge response and axle loads with their accurate positions. Binocular vision technology was used to continuously track the spatial position of the vehicle while it traveled over the bridge. Based on the obtained time–spatial information of the vehicle axles, the ordinate of influence line, axle load, and bridge response were correctly matched in the objective function of the BWIM algorithm. The influence line of the bridge, axle, and gross weight of the vehicle could then be reliably determined. Laboratory experiments were conducted to evaluate the performance of the proposed method. The negative effect of non-constant velocity on the identification result of traditional BWIM methods and the reason were also studied. Results showed that the proposed method predicted bridge influence line and vehicle weight with a much better accuracy than conventional methods under the considered adverse situations, and the stability of BWIM technique also was effectively improved. The proposed method provides a competitive alternative for future traffic load monitoring.

Optimisation of dynamic navigation system for automatic driving vehicle based on binocular vision

Aiming at the problem of low confidence and low success rate of vehicle navigation system, a dynamic navigation system based on binocular vision is proposed and designed. The system is optimised from two aspects: hardware and software. In hardware, binocular vision sensor is added on the basis of traditional system, and the hardware circuit of the system is optimised and modified. In software, accurate road information is obtained by binocular vision technology, and automatic driving route is planned. In the process of automatic driving, the function modules of dynamic driving state detection and vehicle path control navigation, memory and human-computer interaction unit are optimised. The experimental results show that the optimisation system has a 39% improvement in the success rate compared to the traditional navigation system, and the confidence level is improved by 8.06%. It is a navigation system having higher function to achieve success rate and confidence.

Research on Sugarcane Seed-Bud Location Based on Anisotropic Scaling Transformation

Highlights A method of locating sugarcane seed bud based on anisotropy transformation is proposed.Using computer binocular vision technology, the location of the sugarcane seed bud was determined by edge feature matching of the sugarcane seed bud.There are few methods to study the automatic location of sugarcane seed buds, and our research provides a new research idea.Abstract. Sugarcane is a major economic crop in China, but the degree of mechanization in sugarcane cultivation is low. To improve the economic benefit of sugarcane planting, promoting the use of mechanization in sugarcane planting is necessary. Currently, the sugarcane planted using mechanization has a low survival rate and the mechanization efficiency is low because the existing sugarcane precutting machine fails to address the problem of damaging seed buds. This study proposed a sugarcane bud localization method based on computer binocular vision technology. The sugarcane stem segment positions can be determined by the grayscale horizontal projection after preprocessing the sugarcane image based on color and grayscale features. Then, the bud area can be intercepted according to the positional relationship between the seed bud and the stem node, and the planar position of the seed bud will be determined by using the color space conversion and the gray vertical projection. Finally, the anisotropic scaling transformation is used to match the seed-bud area and restore the spatial coordinates of the seed bud, and the spatial position of the seed bud can be determined. The image pyramid acceleration matching process is adopted, which can make the method more suitable for real-time applications. The experimental results show that the accuracy of seed-bud matching based on the anisotropic scaling transformation is 98%, which provides a basis for research on the anti-injury germ system in the automatic planting process of sugarcane. Keywords: Anisotropic scaling, Binocular vision, Image pyramid, Mechanization planting of sugarcane, Seed bud location.

Application of Binocular Stereo Vision in Radioactive Source Image Reconstruction and Multimodal Imaging Fusion

Current methods for the imaging of radioactive sources ignore the surrounding scene information. Reconstruction results do not have real scale meaning, creating difficulties for further disposal and management of nuclear waste. In addition, the traditional reconstruction method requires fixed hardware to directly measure the reconstruction parameters. In recent years, the combination of computer vision and traditional radiation imaging has been greatly performed. The binocular camera is a small, light-insusceptible visual instrument, suitable for integration with other instruments. In this article, we present a method based on binocular vision technology to radioactive source image reconstruction. The binocular camera provides the assisting information about the surrounding scene and replaces the fixed hardware to indirectly achieve the measurement target of reconstruction parameters, which includes an azimuth angle and a distance. In addition, we designed a simple algorithm based on the Hough transform and the regular shape of the waste drum to improve the spatial resolution for good visualization. The experimental rendering results with different transparency degrees provide a good sense of spatial scale.

A Ship Ranging and Positioning System Based on Binocular Vision Technology

Relative ranging and positioning is the basis to ensure the navigation safety of ships, and it has a broad application prospect in various maritime activities, especially in the aspects of docking and undocking, sailing in crowded areas, towing, convoy sailing, and warship replenishment, etc. In order to improve the autonomy and automation of ship ranging and positioning and reduce the cost, a novel type of ship ranging and positioning system based on binocular vision technology is proposed. The proposed system includes seven modules, i.e. video information acquisition module, ship attitude acquisition module, camera parameter calibration module, target recognition and segmentation module, image matching module, ranging and positioning module, and information display module. After acquiring the target image sequences by using two cameras, the proposed system identifies and segments maritime targets quickly by using Canny algorithm in the artificial interactive mode, realizes reduced-space search by using epipolar constraint method, achieve fast target features matching to obtain the vision disparity of the feature points by using the SURF algorithm, and calculates the distance of the target according to the binocular vision ranging principle, so as to realize interactive visual ranging and positioning. The proposed system has the advantages of low cost, simple structure and convenient operation. It can realize relative ranging and positioning with only one computer and two cameras. Experiments demonstrate that the proposed system, with fast running speed and high accuracy, can satisfy the relative ranging and positioning needs of ships.

Accuracy of a Novel Virtual Articulator for Recording Three-Dimensional Dentition.

To research and develop a novel virtual articulator system (the PN-300) based on computer binocular vision, raster scanning, and simulation technology and to conduct a preliminary evaluation of its accuracy. Two digital cameras were used to build the trajectory-tracking part of the virtual articulator system, and cameras combined with a projection module were used to form the scanning part of the system. The most prominent feature of the PN-300 is its ability to simultaneously obtain the 3D data of the subject's teeth and the movement trajectory of the mandible relative to the maxilla. The PN-300 recorded the linear, circular, and rectangular quadrilateral movements of a high-accuracy 3D electronic translation stage. The accuracy of measurement of the inclination of incisal guidance derived from the PN-300 based on the PROTAR evo7 articulator was also estimated. The measurement error was below 100 μm for the linear and circular movements, and the angle error was within 0.2 degrees for the rectangular quadrilateral movements. The error of inclination of protrusive incisal guidance was 1.51 ± 0.68 degrees, and for incisal guidance was 0.82 ± 0.55 degrees. Trajectories and incisal 3D data obtained by the PN-300 were combined with data from plaster models and CBCT to simulate mandibular movement and to calculate the trajectories of the condyle. The PN-300 achieved a good accuracy for recording mandibular movement and can be expected to calculate the movement of the condyle.

Simultaneous localization and mapping of medical burn areas based on binocular vision and capsule networks

Accurate evaluation of burn degree is a key step in the treatment of burn patients. The body surface area of burn area is the main basis to evaluate the degree of burn. To estimate the burn area timely and accurately is the basis of providing correct infusion volume for patients and determining further treatment measures. Therefore, it is necessary to study a fast and effective method to calculate the area of human body burn. For large-area burn patients, accurate fluid replenishment in shock period plays an important role in the maintenance of vital signs and wound healing, and the estimation of burn body surface area is the basis for calculating fluid replenishment in shock period. As an important branch of computer vision, binocular stereo vision has penetrated into many fields of production and life, which is a hot topic in computer application. Binocular stereo vision technology is based on the theory of parallax, which uses binocular camera to collect the left and right views of the measured objects. The paper proposes the binocular vision uses stereo matching algorithm to calculate the position deviation between two images, so as to obtain the 3D geometric information of the object. Based on this, this paper uses binocular vision technology and capsule network model to build a medical burn area evaluation model. The experimental results show that the method proposed in this paper can effectively locate the burn area and image processing.

Three-dimensional reconstruction of the dribble track of soccer robot based on heterogeneous binocular vision

With the rapid development of science and technology, as an important branch of the computer. Binocular vision technology has also been greatly improved and improved. It has been applied to many fields of production and life, medical information, industrial manufacturing, and service industries. Heterogeneous binocular vision typically uses a camera to simulate the sensory acquisition process of humans. To process the target information, multiple viewpoints observe the same motion scene. Finally, multiple pairs of images are obtained from different viewpoints. Three-dimensional reconstruction has long been one of the important research fields of computer vision. It has great application value in robot autonomous navigation vision, self-driving vehicle, and automatic detection field and freedom degree mechanical device control. In the robot application of visual positioning technology, with the development needs, the level of intelligence of the robot is also gradually increasing. The three-dimensional reconstruction technology combined with the principle of heterogeneous binocular vision has important practical significance for completing the technique of the dribble track of soccer robots. Based on the RoboCup soccer robot competition, this paper uses the heterogeneous binocular vision technology to analyze the kinematics of the soccer robot in detail. By analyzing the basic motion characteristics of soccer robots, a three-dimensional reconstruction technology model of its dribble trajectory is established, and the reliability, practicability and rationality of the design are verified.

Obstacle visual sensing based on deep learning for low-altitude small unmanned aerial vehicles

An obstacle real-time sensing method, which is based on deep learning and target tracking method and integrated with monocular vision and binocular vision, is proposed for unmanned aerial vehicles (UAVs) in this paper. Firstly, it uses the deep learning method to detect and recognize the first-frame figure collected by cameras. Then, it uses the target tracking algorithm to track the detection results for the first-frame figure in real time to improve the real-time performance of the detection system. Meanwhile, it uses the binocular vision technology to execute the three-dimensional reconstruction for the current frame of the entire figure to obtain the environmental spatial information. Subsequently, combined with the points clustering strategy and the information fusion method, it can resolve the types, spatial locations, and outlines of obstacles. Finally, to verify the proposed method, we developed a physical prototype, and the results showed that the real-time sensing for obstacles can be realized under the condition that UAVs are equipped with one binocular camera.

Accuracy Improvement of Binocular Vision Measurement System for Slope Deformation Monitoring.

This paper studies the limitations of binocular vision technology in monitoring accuracy. The factors affecting the surface displacement monitoring of the slope are analyzed mainly from system structure parameters and environment parameters. Based on the error analysis theory, the functional relationship between the structure parameters and the monitoring error is studied. The error distribution curve is obtained through laboratory testing and sensitivity analysis, and parameter selection criteria are proposed. Corresponding image optimization methods are designed according to the error distribution curve of the environment parameters, and a large number of tests proved that the methods effectively improved the measurement accuracy of slope deformation monitoring. Finally, the reliability and accuracy of the proposed system and method are verified by displacement measurement of a slope on site.