A MultiScale Particle Filter Framework for Contour Detection.

The object contour detection in natural scenes is an important research problem in computer vision. The difficulty is that the integrity of contour extracted is interfered with a large number of texture edges in the background seriously. In recent years, some researchers have introduced image contour detection method with biological visual features and achieved preferable results. Such as, based on the fact of visual outer region suppression, they can suppress a certain amount of texture edge while extracting the contour of the object, and obtain the proper contour edge. However, when this method is used for contour detection in some image which the texture is similar to the contour, the texture edge still cannot be removed well, and the result is unsatisfactory. In consideration of the merit and demerit of the feature suppressed method, we propose an improved algorithm for object contour detection in natural scene based on image saliency. Firstly, the initial contour is obtained by using the suppression effect of non-classical receptive field, and the saliency map is obtained by using the hyper-complex Fourier transform image saliency detection method, and then the initial contour and the saliency map are integrated into the final contour. Compared with the traditional contour detection method, the contour extraction algorithm based on image saliency can remove more background information and improve the accuracy and integrity of object contour extracted in natural scenes.

Online detection of yarn roll’s margin is one of the key issues in textile automation, which is related to the speed and scheduling of bobbin (empty yarn roll) replacement. The actual industrial site is characterized by uneven lighting, restricted shooting angles, diverse yarn colors and cylinder yarn types, and complex backgrounds. Due to the above characteristics, the neural network detection error is large, and the contour detection extraction edge accuracy is low. In this paper, an improved neural network algorithm is proposed, and the improved Yolo algorithm and the contour detection algorithm are integrated. First, the image is entered in the Yolo model to detect each yarn roll and its dimensions; second, the contour and dimensions of each yarn roll are accurately detected based on Yolo; third, the diameter of the yarn rolls detected by Yolo and the contour detection algorithm are fused, and then the length of the yarn rolls and the edges of the yarn rolls are calculated as measurements; finally, in order to completely eliminate the error detection, the yarn consumption speed is used to estimate the residual yarn volume and the measured and estimated values are fused using a Kalman filter. This method overcomes the effects of complex backgrounds and illumination while being applicable to different types of yarn rolls. It is experimentally verified that the average measurement error of the cylinder yarn diameter is less than 8.6 mm, and the measurement error of the cylinder yarn length does not exceed 3 cm.

In Cassini ISS (Imaging Science Subsystem) images, contour detection is often performed on disk-resolved objects to accurately locate their center. Thus, contour detection is a key problem. Traditional edge detection methods, such as Canny and Roberts, often extract the contour with too much interior details and noise. Although the deep convolutional neural network has been applied successfully in many image tasks, such as classification and object detection, it needs more time and computer resources. In this paper, a contour detection algorithm based on H-ELM (Hierarchical Extreme Learning Machine) and DenseCRF (Dense Conditional Random Field) is proposed for Cassini ISS images. The experimental results show that this algorithm’s performance is better than both traditional machine learning methods, such as Support Vector Machine, Extreme Learning Machine and even deep Convolutional Neural Network. The extracted contour is closer to the actual contour. Moreover, it can be trained and tested quickly on the general configuration of PC, and thus can be applied to contour detection for Cassini ISS images.

A PC-based system for real-time left ventricular endocardial contour detection on echocardiographic images has been developed in our laboratories. The system consists of an IBM-compatible PC equipped with a commercially available frame grabber and the special-purpose Minimum Cost Contour Detector (MCCD) board developed for this project. The MCCD board uses 1 or 2 Intel i960 processors and receives digital images (512*512*8 bit) from the frame grabber over a digital video bar in real time. The contour detection algorithm is based on dynamic programming. Detected contours and other results are transferred to the host via the AT-bus. Processing time is short enough for real-time performance in a practical situation. Area of contours and regional areas in 8 segments are calculated in real time and displayed in graphs on the host system. The MCCD board is fully programmable and may be used for similar real-time contour detection applications, e.g. for X-ray angiography and intravascular ultrasound. >

A semi-automatic endocardial contour detection method has been developed on cine CT sequences. A preprocessing relying on a pyramidal thresholding is made, permitting the extraction of a region of interest on the first image. A contour detection algorithm is then processed on the whole sequence, grounded on the search of an optimal pathway in the graph.

Many cues have been proposed for contour detection or image segmentation. These include low-level image gradients to high-level information such as the identity of the objects in the scene or 3D depth understanding. While state-of-the-art approaches have been incorporating more cues, the relative importance of the cues is unclear. In this paper, we examine the relative importance of low-, mid- and high-level cues to gain a better understanding of their role in detecting object contours in an image. To accomplish this task, we conduct numerous human studies and compare their performance to several popular segmentation and contour detection machine approaches. Our findings suggest that the current state-of-the-art contour detection algorithms perform as well as humans using low-level cues. We also find evidence that the recognition of objects, but not occlusion information, leads to improved human performance. Moreover, when objects are recognized by humans, their contour detection performance increases over current machine algorithms. Finally, mid-level cues appear to offer a larger performance boost than high-level cues such as recognition.

Contour detection via stacking random forest learning

Existing buildings often lack construction or as-built drawings, leading to a lack of references during supplementary property registration or design adjustments. Therefore, it is necessary to carry out as-built surveying and produce architectural drawings. Traditional methods such as total stations are often used, but they are inefficient, have poor visualization, and incomplete data. This study employs a handheld 3D laser scanner and integrates laser point clouds with oblique photogrammetry to establish a comprehensive workflow for as-built architectural surveying and mapping, spanning from field measurement to office mapping. A contour detection and regularization algorithm based on prior constraints is proposed to achieve automated extraction of window contours. Through practical validation, the results show a mean error of less than 5 cm, improved project efficiency, strong operability, and significant potential for widespread application.

Contour detection is an important step in contour based object analysis in 2D images. We present a novel method for contour detection optimized for contour traversal tasks in 2D images. The method is based on morphological operations and detects a one-pixel wide contour around objects in binary images. Performance of the method is evaluated on ground truth images from our dataset created by manually tracing the contour in images. Proposed method achieves high DICE coefficient overlap to ground truth images and superior detection compared to other state-of-the-art methods.

Eyelid contour detection and tracking for startle research related eye-blink measurements from high-speed video records

A very accurate detection is required for fitting 3D dental model onto color images for tracking the milimetric displacement of each tooth along orthodontics treatment. Detecting the teeth boundaries with high accuracy on these images is a challenging task because of the various quality and high resolution of images. By training Mask R-CNN on a very large dataset of 170k images of patients’ mouth taken with different mobile devices, we have a reliable teeth instance segmentation, but each tooth boundaries are not accurate enough for dental care monitoring. To address this problem, we propose an efficient method for object contour refinement using instance segmentation (CRIS). Instance segmentation provides high-level information on the location and the shape of the object to guide and refine locally the contour detection process. We evaluate CRIS method on a large dataset of 600 dental images. Our method improves significantly the efficiency of several state-of-the-art contour detectors: Canny (+32.0% in ODS F-score), gPb (+17.8%), Sketch Tokens (+17.3%), Structured Edge (+12.2%), DeepContour (+15.5%), HED (+2.9%), CEDN (+2.2%), RCF (+2.2%) and also the best result (ODS F-score of 0.819). Our CRIS method can be used with any contour detection algorithms to refine object contours. In that way, this approach is promising for other applications requiring very accurate contour detection.

We apply a cellular neural network (CNN) based contour detection to exact the main reflective interface in seismic image. The seismic image is generated by reverse time migration (RTM) using the pseudospectral time domain (PSTD) method. According to Nyquist sampling theorem, the PSTD algorithm requires only two points per minimum wavelength rather than the traditional high order finite difference time domain (FDTD) which needs more than eight points per minimum wavelength to get the same accuracy. Thus, RTM using the PSTD algorithm can reduce the computing costs greatly when comparing with the traditional RTM based on the FDTD algorithm. To get more clear reflective interface in imaging result of reverse time migration, we use a contour detection algorithm based on CNN which calculates the template parameters by the gray-scale and spatial relationship between the central pixel and the other neighboring pixels in the current local window. The simulation results shows that the proposed method have good efficiency and imaging quality, and the contour detection method makes the reflective interface easier to distinguish.

The paper presents an oval detection algorithm in color images. First a SOBEL operator is used to do edge detection and enhancement. Then Nilsson's algorithm for contour detection is applied with small changes. To enhance robustness against gaps in possible contours, adjacency range is increased. As a result, sets of potential contours are obtained. As an alternative the authors introduce their new approach which first identifies regions of potential ellipses, discarding all that do not fit the aspect ration threshold set up by the authors. Then each potential region is split in three segments. For each segment the mean gradient is calculated and a possible oval is identified if the mean gradient is increasing or decreasing from the first to the third region. Once a potential oval is identified, ten point couples with the largest distances are used to determine the major axis. The deflection angle for the minor axis is computed and two points are found nearest to that line so at the end the five points of a potential ellipse are found. Two indicators are now used to reflect similarity and approach to do best possible ellipse fitting.

As the high voltage circuit breaker works, the metal parts are prone to many types of mechanical defects, such as plastic deformation, metal loss, corrosion, and so on. The defects will inevitably decrease the performance, mechanical reliability, lifetime, and other critical parameters of the breaker. Therefore, the detection of mechanical defects is of great help to reduce the mechanical failure rate of circuit breakers. In the present paper, a novel method based on edge detection and deep learning is proposed to detect the mechanical defects of the high voltage circuit breaker. Firstly, the high resolution images of the circuit breaker are photographed, and the components in the images are segmented in the minimum range by the contour detection algorithm. After binarization and morphological processing of the segmented images, the edge is drawn with the improved edge detection algorithm. However, the components of the circuit breaker are complicated, which results that the edge features of the critical component are difficult to extract. Then, the specific contour of components is detected by the depth estimation and segmentation methods. Secondly, a deep learning platform base on Tensorflow is established, which is an optimization method based on Convolutional Neural Network. In order to enhance the feature extraction capacity, a convolution kernel is inputted into a sparse self-coding network to proceeds with the optimal pre-training. Based on the above method, the features of images are automatically learned, and each convolution carries for the equilibrium of image entropy of convolution kernel by introducing similarity constraint rule. Finally, 3 types of mechanical defects, including plastic deformation, metal loss, corrosion, images and non-defect images as input are exploited to validate the CNN with convolution kernel optimized. The experimental results show that the average accuracy of the approach is better than the traditional convolution neural network model and has good feature extraction ability and generalization ability. In conclusion, combined with the improved edge detection and deep learning algorithms, the detection of typical mechanical defects can be well performed. For the collected images of circuit breaker, so as to reduce the impact of noise on defect accuracy of mechanical, the proposed method provides a reference for the non-destructive mechanical defect detection of high voltage circuit breaker.

Abstract. In this study, we present a novel method of identifying and geolocalizing cloud field elements from a portable all-sky camera stereo network based on the ground and oriented towards zenith. The methodology is mainly based on stereophotogrammetry which is a 3-D reconstruction technique based on triangulation from corresponding stereo pixels in rectified images. In cases where clouds are horizontally separated, identifying individual positions is performed with segmentation techniques based on hue filtering and contour detection algorithms. Macroscopic cloud field characteristics such as cloud layer base heights and velocity fields are also deduced. In addition, the methodology is fitted to the context of measurement campaigns which impose simplicity of implementation, auto-calibration, and portability. Camera internal geometry models are achieved a priori in the laboratory and validated to ensure a certain accuracy in the peripheral parts of the all-sky image. Then, stereophotogrammetry with dense 3-D reconstruction is applied with cameras spaced 150 m apart for two validation cases. The first validation case is carried out with cumulus clouds having a cloud base height at 1500 m a.g.l. The second validation case is carried out with two cloud layers: a cumulus fractus layer with a base height at 1000 m a.g.l. and an altocumulus stratiformis layer with a base height of 2300 m a.g.l. Velocity fields at cloud base are computed by tracking image rectangular patterns through successive shots. The height uncertainty is estimated by comparison with a Vaisala CL31 ceilometer located on the site. The uncertainty on the horizontal coordinates and on the velocity field are theoretically quantified by using the experimental uncertainties of the cloud base height and camera orientation. In the first cumulus case, segmentation of the image is performed to identify individuals clouds in the cloud field and determine the horizontal positions of the cloud centers.