Search

There are variety of fruits around the world, different types of fruits contain different types of nutrients and vitamins which could benefits our health. In order to understand which fruit can provide specific type of nutrients, we need to identify the types of fruits. However, fruits grow in a different shape, colour and texture based on the country they were planted and the environment of the land. Implementing a machine vision-based recognition on the fruits can help people recognize them easily. In this paper, an edge detection method is applied using computer vision approach to recognize different types of fruits. The fruits are classified based on the features extracted from their images. In the experiment, a total of 450 images of three types of fruit are used, which are apples, lemons and mangoes. Pre-processing steps are applied on the captured image to improve the quality of fruit details and the edge features are extracted using Canny Edge Detection method. Classification of the fruits is accomplished using two different types of learning model, the deep leaning model, Convolution Neural Network (CNN) and machine learning model, Support Vector Machines (SVM). The performance of both classifiers is compared and the model with the best performance, SVM is chosen as the model for the system. The system can achieve 86% classification accuracy with the SVM model, which is good enough for fruit recognition.

In this paper, an effective and simple Grid based vanishing point detection position estimation algorithm and Enhanced GIST descriptors based indoor scene recognition algorithm for navigation of MAV in indoor corridor environment is described. Two different classifiers, k-nearest neighbour classifier and support vector machine is employed for the categorization of indoor scenes into corridor, staircase or room. Indoor scene classification was performed on Dartaset-1. In the training phase of the indoor scene recognition algorithm, GIST, HODMG and Enhanced-GIST feature vectors are extracted for all the indoor training images in the Dataset-1 and indoor scene classifiers are trained for the extracted image feature vectors and assigned image labels of the indoor scenes (corridor-1, staircase-2 and room-3). In the testing phase of the indoor scene recognition algorithm, for each unknown test image frame GIST, HODMG and Enhanced-GIST feature vectors are extracted and the indoor scene classification is performed using a trained scene recognition model. The proposed indoor scene recognition algorithm using SVM with Enhanced GIST descriptors produced high recognition rates of 99.33% compared to the KNN classifiers. After recognizing the indoor scene as corridor, the MAV has to estimate its position based on the detection of vanishing point in the indoor corridor image frames. Experimental results show that the proposed method is suitable for real time operations.

In recent years, computer vision technology in the field of science and technology has been stubbornly high, making a lot of scientific and technical personnel involved in the research gate technology. Depth learning, artificial intelligence in recent years, often around our ears, vision technology is occupying a place in the field of artificial intelligence. This paper mainly studies the recognition and localization of robots based on vision. The mobile robot is detected by the background difference method, and the background model is established by the Gaussian background modeling. Recognizing the color and number tags on the robot for matching the robot’s features. Real-time positioning of the robot under the premise of the same background.

In this paper a vision based detection method for humping in high-speed laser welding is proposed. A high-speed imaging system is used to capture the images of melt pool in the near-infrared range and the unaffected melt pool region is extracted as the ROI (region of interest) according to the structure of images. The occurrence of humping and positions of humps are determined according to the geometric characteristics of the melt pool in the ROI. Experimental results indicate that the proposed method can detect positions of humps when humping occurs and can also avoid false detection when no humping occurs. The potential of the proposed method is demonstrated for humping control.

Based on computer digital technology, the paper first uses the method of non-equal interval quantization to quantize the pixels in the 3D image, and introduces the artificial intelligence vision technology HSV colour space to construct the feature vector formula after non-equal interval quantization to calculate the difference between the two images the final distance between. Experiments have shown that the algorithm has better robustness to changes in light intensity while maintaining the lower number of quantized bins in the Tahoun algorithm. At the same time, the algorithm can effectively reduce 3D colours, with better visual effects, and better than the original. The interference of images with large differences on the search results improves the recall rate of the image library used in the experiment.

This paper describes a system that can distinguish between visually-similar objects, specifically bicycles and motorcycles, successfully from the vantage point of traffic surveillance cameras. The You Only Look Once (YOLO) is used as the main framework in this research due to its speed performance among various machine learning models and methods evaluated. We built a dataset consisting of motorcycle and bicycle images from different CCTV footage for this project. CCTV footage may vary on the angle of view from the object, image resolution, and ambient environment settings. Using this dataset, we trained YOLOv3-based models, and their performances were compared to the vanilla version of YOLOv3 and other pre-trained models. Four (4) models were trained and compared; the best-performing model is shown to be associated with a dataset with properly labeled data (i.e., marking every instance of the object of interest) and having the most number of instances in the training and testing set.Introduction.

Aiming at the low efficiency and low accuracy of traditional manual detection of workpiece circular features, a machine vision method is proposed to accurately detect circular features. By adopting algorithms such as iterative threshold segmentation, generating XLD contours, selecting feature contours and random Hough transforms, single or multiple circles on the workpiece can be quickly detected. Experiments show that the detection is efficient and accurate, and can meet the detection needs of industry.

Several prototypes of mechanical fingers composing anthropomorphic prosthetic hands have been developed using underactuated mechanisms. These mechanisms require specific experimental procedure to measure their properties. Due to their low weight, the addition of further sensors for measurement purposes could modify the trajectory of the single parts and the overall prosthetic device behaviour. Vision system methods represent a route to overcome these drawbacks. In this study, an experimental methodology to acquire the dynamics of an underactuated prosthetic finger is presented due to the employment of a vision-based technique, the Digital Image Correlation. The experimental characterization aids the development of a dynamical model of the finger behaviour.

Anthropometry can be defined as the science of dimensional measurement of the size and proportions of the human body. Anthropometric measurement is used in a lot of fields such as medical, forensics and clothing. The modern anthropometry techniques are expensive and takes up a lot of space while traditional anthropometry techniques are slow and susceptible to human error. Hence, there is a need for the development of a vision-based estimation system that is cheap, fast, accurate and portable. The study in this project proposes a noble way of using Raspberry 3B+ with Logitech C310 camera to perform estimation of chest, waist, hip circumference and body height through 2D images. The system works by implementing Pixel Density Method, HSV thresholding method, OpenPose deep neural network, bounding box method, pixel counting method and Ramanujans ellipse circumference approximation method to achieve objectives of this study. The percentage error between the estimated result and the measured result of chest, waist, hip circumference and body height are only 2.11%, 4.66%, 4.31% and 1.74%. In conclusion, a vision based anthropometric estimation system is successfully developed with high accuracy using all the proposed methods.

Recently, Lane departure warning system has attracted great attention as it contributes to vehicle active safety. In this paper, a vision-based lane departure warning system under high-speed driving is proposed. The system consists of two functional parts: lane markings detection and vehicle departure identification. The Hough Transform is applied to detect lane boundaries, which is a most effective detection method with high reliability. Based on the road line, a method using several Euclidean-distance-related parameters to calculate vehicle’s position and its deviating status is proposed, which addresses the problem of efficient detection of lane departure under high-speed driving condition. The algorithm is tested and verified in various real driving conditions and proved a reliable and steady performance.

Hazard detection is the key technique for autonomous landing during a planetary exploration mission. This paper proposed an end-to-end spatiotemporal network to detect all hazards in the image sequences captured by optical camera. The spatial stream processes the colour image sequences and the temporal one learns features from optical flow image sequences. The spatial features and temporal features are fused by the proposed metric fusion method, and then the spatiotemporal features become more discriminative through triplet loss. A hazard map can ultimately be obtained in the testing phase after processing on the full-size image sequences. The evaluation results prove the effectiveness of the proposed network, and the testing results show the feasibility of practical application.

With the advent of digital imaging and image processing techniques vision based monitoring and characterisation of combustion flames have developed rapidly in recent years. This paper presents a short review of the latest developments in this area. The techniques covered in this review are classified into two main categories: two-dimensional (2D) and 3D imaging techniques. Experimental results obtained on both laboratory- and industrial-scale combustion rigs are presented. Future developments in this area also included.

This paper reviews the vision-based localization and map construction methods from the perspectives of VSLAM, SFM, 3DMax and Unity3D. It focuses on the key technologies and the latest research progress on each aspect, analyzes the advantages and disadvantages of each method, illustrates their implementation process and system framework, and further discusses the way to promote the combination for their complementary strength. Finally, the future opportunity of the combination of the four techniques is expected.

In this paper, a vision-based method for measuring a civil engineering construction's in-plane deflection curves is presented. The displacement field of the analyzed object which results from loads was computed by means of a digital image correlation coefficient. Image registration techniques were introduced to increase the flexibility of the method. The application of homography mapping enabled the deflection field to be computed from two images of the structure, acquired from two different points in space. An automatic shape filter and a corner detector were implemented to calculate the homography mapping between the two views. The developed methodology, created architecture and the capabilities of software tools, as well as experimental results obtained from tests made on a lab set-up and civil engineering constructions, are discussed.