Perform Edge Detection Research Articles

Noise-Robust, Reconfigurable Canny Edge Detection and its Hardware Realization

Edge detection is a common operation in image/video processing applications. Canny edge detection, which performs well in different conditions, is one of the most popular and widely used of these algorithms. Canny’s superior performance is due mainly to its provision of the ability to adjust the output quality by manipulating the edge detection parameters, Sigma and Threshold. Calculating values for these two parameters on-the-fly and based on the application’s circumstances requires additional preprocessing, which increases the algorithm’s computational complexity. To reduce the complexity, several proposed methods simply employ precalculated, fixed values for the Canny parameters (based on either the worst or typical conditions), which sacrifices the edge detection’s performance in favor of the computational complexity. In this paper, an adaptive parameter selection method is proposed that selects values for the Canny parameters from a configuration table (rather than calculating in run-time), based on the estimated noise intensity of the input image and the minimum output performance that can satisfy the application requirements. This adaptive implementation of the Canny algorithm ensures that, while the edge detection performance (noise robustness) is higher than state-of-the-art counterparts in different circumstances, the execution time of the proposed Canny remains lower than those of recent cutting-edge Canny realizations.

Edge detection methods and key parameters study

Edge is essential for diverse vision processing tasks and it constitutes a crucial part in the image of objects. There has already been a huge amount of researches about edge detection algorithms. The Canny method and the Sobel method are the most widely used techniques because of their relative advantages. Some parameters can drastically affect the performance and results of the two methods, which will be discussed in this paper. In addition, noise is also a factor affecting the number of the extracted edges. Experiment results have demonstrated that the noise filter will achieve better edge detection performance. Further improvements with suitable parameters of different applications for the two methods are possible.

Intensity gradient based edge detection for pixelated communication systems

Pixelated optical communication systems transmit data using a series of encoded pixelated images. The decoding of these pixelated images is important for reliable data communication. This study proposes a novel intensity gradient-based edge detection (IGED) method for the received pixelated images. In the first stage of the proposed edge detection algorithm, the intensity difference of neighbouring pixels is calculated in the horizontal or vertical direction depending on the presence of a horizontal or vertical edge. In the second stage, the false edge pixels whose intensity values are less than the threshold are removed. Next, slopes are calculated between the consecutive edge pixels and the appropriate lines are found considering the slopes that have almost the same value. The new IGED method performs edge detection without requiring any anchor pixel blocks. This study also provides a description of a misalignment correction algorithm suitable for pixelated images. Next, the effectiveness of the proposed IGED method is tested through an experimental demonstration. It is shown that the IGED method can properly detect the edges even for pixelated images affected by rotational misalignment. For a given transmitter resolution, the proposed method provides better data rates than the existing methods.

Reconstruction of Inner Worm Passage of Quercus Mongolica Seeds

This paper takes Quercus mongolica as an example, In order to determine the position of insects inside the Quercus mongolica seed. The seeds of Quercus mongolica were cut at a certain thickness to obtain an internal cross-sectional image of the seeds. In this paper, Canny and Hough algorithm are used to perform edge detection and line detection on unprocessed images, and determine the coordinate origin of the study area so that all study areas are in the same coordinate range. In order to obtain the position information of the insect path, based on the RGB principle, this paper designs a screening algorithm for the pixel points in the study area, and obtains the position information of the insect track. In order to observe the insect path more intuitively, this paper establishes a three-dimensional data model, and uses MATLAB to draw a three-dimensional structure diagram of the position information of the insect path, and shields the interference items such as seed germ to obtain a more accurate movement track of the insect.

Automatic prior shape selection for image edge detection with modified Mumford–Shah model

Edge detection plays an important role in the field of image processing. In this paper, we propose a novel variational model to automatically and adaptively detect one or more prior shapes from the given dictionary to guide the edge detection process. In that way, we can effectively detect the shapes of interest from the test image. Moreover, an efficient algorithm based on the Alternating Direction Method of Multipliers (ADMM) is proposed to solve this model with guaranteed convergence. A variety of numerical experiments show that the proposed method has achieved ideal performance for edge detection in images with missing information, various types of noise and complicated background, and even multiple objects.

Application of Edge Preserving and Interpolation Algorithm Based on Wavelet Transformation in Mechanical Inspection and Flaw Detection

In mechanical inspection and flaw detection, the fault point in the image is often small, it is easy to cause the occurrence of missed detection, and due to the internal factors of the imaging system and environmental effects, the final image is often degraded, which is manifested as the deterioration of subjective vision and the decrease of actual resolution. In order to improve the resolution of the image without changing the existing imaging system and prevent the occurrence of the phenomenon of missed detection. An interpolation edge preserving algorithm based on wavelet transformation is proposed, the algorithm firstly performs edge detection and interpolation for low-resolution image, after that, the initial interpolated magnified image was decomposed by wavelet, and the low-frequency component image after wavelet decomposition was replaced by the original low-resolution image. Then, the inverse wavelet transform was carried out to obtain the high-resolution image with good edge information. The experimental results showed that compared with the traditional interpolation method and wavelet interpolation method, this algorithm performs better in the edge visual effect and objective index, and it is an effective algorithm to improve the resolution of detection image.

Determination Of Errors On The Surface Of The Packaging By Image Processing Technique

This work; The aim of this study is to detect the defects on the surfaces of packaging products by means of Image Processing in a company producing food packaging.In the Images of products received from the company, with the support of OpenCV Library in Python Programming Language in Pycharm Platform; types of defects such as paint, signal, misalignment, knife line are detected. The OpenCv Library saves time and improves system performance. In this study, firstly by color masking, two sets of blue color which are “dusuk” and “yuksek” are created. The image is converted to HSV format, because the masking process requires the HSV format of the image. After masking, the blue background becomes white and all other colors become black. After this process, the pixels with white color are converted to their original values so that the values of the pixels with black color do not change. In order to perform Edge Detection, the image is converted to gray tone so that unwanted noise is obtained in the image. Then, the images of the faulty parts are taken and similarities are applied to find the faulty parts in the real image. Finally, the defective parts are indicated in the yellow rectangle. Operators in the company make the error detection by visual inspection. In this system, when the operator cannot detect the error, the wrong packages continue to be produced in the coils where the wrapping is made. In order to minimize the costs incurred in this situation, it was decided to use the Image Processing method and appropriate results were obtained.

Vision-based edge detection between plant column and soil of ningxia lycium barbarum garden

ABSTRACTThe cultivation of Lycium barbarum (L.barbarum) is a highly traditional and advantageous industry in Ningxia, China, and has strong development prospects. At present, the protection, fertilization, picking, and other production aspects of L.barbarum are generally inefficient and labor intensive, presenting a bottleneck that restricts the development of the industry. Developing intelligent production equipment in the form of a ‘general self-moving host platform + operation module’ is an urgent task for the healthy development of the L.barbarum industry. A self-planning, self-organizing, host platform must be able to perform adaptive navigation in complex unstructured environments. For this purpose, a method of edge detection that can distinguish between the plant column and soil is required. Using a color difference model with Otsu’s method for image segmentation, a corrected gradient image based on the marking method is employed to remove small noise regions and then perform edge detection. Experiments demonstrate that one particular color model offers strong adaptability for light and shadow, which is good for distinguishing between the plant column and soil. The proposed method can effectively detect the edges between the plant column and soil, laying the foundation for detecting a suitable path for a self-moving platform and visual navigation.

Processing of Drone’s Digital Image for Determining Border of Rice Fields with Edge Detection Method

Edge detection was the basic thing used in most image processing applications to get information from the image frame as a beginning for extracting the features of the segmentation object that will be detected. Nowadays, many edge detection methods create doubts in choosing the right edge detection method and according to image conditions. Based on the problems, a study was conducted to compare the performance of edge detection using methods of canny, Sobel and laplacian by using object of rice field. The program was created by using the Python programming language on OpenCV. The result of the study on one image test that the Canny method produces thin and smooth edges and did not omit the important information on the image while it has required a lot of computing time. Classification is generally started from the data acquisition process; pre-processing and post-processing. Canny edge detection can detect actual edges with minimum error rates and produce optimal image edges. The threshold value obtained from the Canny method was the best and optimal threshold value for each method. The result of a test by comparing the three methods showed that the Canny edge detection method gives better results in determining the rice field boundary, which was 90% compared to Sobel 87% and laplacian 89%.

A 3-Pixel Fuzzy Mask for Edge Detection

A new group of fuzzy masks is proposed in this study to perform edge detection of images. A group of 3-pixel masks is used to detect edges in four directions. The resulting images retain more edges and show better continuity of the edges compared to the ones produced by the conventional 3 x 3 fuzzy masks. This is accomplished more efficiently using fewer fuzzy rules. Furthermore, thinning functions incorporated yielded edges with good continuity and optimal thickness.

Enhancing metabolic event extraction performance with multitask learning concept.

To extract and generate a valid metabolic pathway from research articles, biologists need substantial amounts of time to digest unstructured text. Text mining currently plays a central role in this research area, because it provides the ability to automatically discover useful information in a reasonable time. A text mining model can be built using a training data or a corpus in supervised manner. Unfortunately, a corpus of the domain of interest may not be always available or insufficient in practice, because a corpus construction is a labor-intensive task and needs specialist annotation. In this paper, we developed an event extraction system, a text-mining task, to extract metabolic interactions from research literature and then reconstruct metabolic pathways. The proposed system consists of the pipeline of four supervised-learning steps: named entity recognition, trigger detection, edge detection, and event reconstruction. We also introduced a multitask-learning algorithm, a transfer-learning paradigm, that can leverage additional resources of an existing source domain to facilitate a classification of the metabolic event extraction in the target domain. To demonstrate a proof of concept, edge detection, a core step in our event extraction system, was used as a case study in multitask-learning classification. The experimental results showed that the proposed event extraction system provided competitive performance against those of state-of-the-art related system. In particular, the proposed multitask-learning can improve the performance of edge detection, therefore the overall performance of the event extraction system was also improved accordingly.

Superresolution of Noisy Remotely Sensed Images Through Directional Representations

We develop an algorithm for single-image superresolution of remotely sensed data based on the discrete shearlet transform. The shearlet transform extracts directional features of signals and is known to provide near-optimally sparse representations for a broad class of images. This often leads to superior performance in edge detection and image representation when compared with isotropic frames. We justify the use of shearlets mathematically, before presenting a denoising single-image superresolution algorithm that combines the shearlet transform with sparse mixing estimators (SMEs). Our algorithm is compared with a variety of single-image superresolution methods, including wavelet SME superresolution. Our numerical results demonstrate competitive performance in terms of peak-signal-to-noise ratio and structural similarity index metric.

An approach for parameterized shadowed type-2 fuzzy membership functions applied in control applications

At present time, general type-2 fuzzy logic provides better uncertainty modeling when compared to interval type-2 fuzzy logic, and this has been shown with the improvement in the performance of fuzzy logic controllers and fuzzy edge detectors. However, higher computational cost represents a limitation for many applications and implementation platforms, and it is because of this that the computational cost reduction is the main goal of this paper. The aim of this work is to model a GT2 FLS based on shadowed type-2 fuzzy sets (ST2 FS) and apply this approach to control problems. The ST2 FS consists of approximating the secondary membership function by shadowed sets and is because of this, that we present an analytic approach to obtain the shadowed set for triangular and Gaussian membership functions for any parameters. Based on this, we achieve a representation of the ST2 FS as a parameterized function, facilitating its implementation. The results are compared against the performance when a generalized type-2 fuzzy inference system is approximated using α-planes for control applications.

Comparison of deep convolutional neural networks and edge detectors for image-based crack detection in concrete

This paper compares the performance of common edge detectors and deep convolutional neural networks (DCNN) for image-based crack detection in concrete structures. A dataset of 19 high definition images (3420 sub-images, 319 with cracks and 3101 without) of concrete is analyzed using six common edge detection schemes (Roberts, Prewitt, Sobel, Laplacian of Gaussian, Butterworth, and Gaussian) and using the AlexNet DCNN architecture in fully trained, transfer learning, and classifier modes. The relative performance of each crack detection method is compared here for the first time on a single dataset. Edge detection methods accurately detected 53–79% of cracked pixels, but they produced residual noise in the final binary images. The best of these methods was useful in detecting cracks wider than 0.1 mm. DCNNs were used to label images, and accurately labeled them with 99% accuracy. In transfer learning mode, the network accurately detected about 86% of cracked images. DCNNs also detected much finer cracks than edge detection methods. In fully trained and classifier modes, the network detected cracks wider than 0.08 mm; in transfer learning mode, the network was able to detect cracks wider than 0.04 mm. Computational times for DCNN are shorter than the most efficient edge detection algorithms, not considering the training process. These results show significant promise for future adoption of DCNN methods for image-based damage detection in concrete. To reduce the residual noise, a hybrid method was proposed by combining the DCNN and edge detectors which reduced the noise by a factor of 24.

A Fast and Efficient Approach for Image Compression Using Curvelet Transform

In this paper a novel image compression technique using features of wavelet and curvelet transforms is proposed to improve efficiency and compression performance. Indeed, the curvelet transform is one of the recently developed multiscale transforms which is especially designed to represent efficiently curves and edges in an image. In the proposed method, the compression algorithm involves the Haar wavelet transform to decompose the image into four frequency sub-bands. The lowest frequency sub-band coefficients are processed using Set Partitioning In Hierarchical Trees (SPIHT) encoding. Meanwhile, Fast Discrete Curvelet Transform (FDCT) is applied to the remaining frequency sub-bands. The FDCT output coefficients are then quantized according to the sub-band they belong to. The lowest frequency FDCT output coefficients are quantized using Differential Pulse Code Modulation, the medium frequency coefficients are processed using SPIHT, whereas the high frequency coefficients are removed. Experimental results demonstrate that our method provides high performance for edge detection compared to existing techniques particularly for images with abrupt changes. In addition, this new image coding and decoding approach is powerful in terms of computation time. Moreover, the proposed method reveals significant improvement in compression ratio and decoded peak-signal-to-noise-ratio.

Two-Dimensional Edge Detection by Guided Mode Resonant Metasurface

In this letter, a new approach to perform edge detection is presented using an all-dielectric CMOS-compatible metasurface. The design is based on guided-mode resonance which provides a high quality factor resonance to make the edge detection experimentally realizable. The proposed structure that is easy to fabricate, can be exploited for detection of edges in two dimensions due to its symmetry. Also, the trade-off between gain and resolution of edge detection is discussed which can be adjusted by appropriate design parameters. The proposed edge detector has also the potential to be used in ultrafast analog computing and image processing.

Location of wireless sensor networks based on port management system

At present, location is currently the most popular technique for accurate localization in port. The signal is vulnerable to the interference of noise and other wireless systems, so that new requirements are putted forward in the wireless sensor networks between-ships and between-ports. This paper presents a method to mitigate the interference and estimate the location in the wireless sensor networks based on port management system. At first, independent component analysis (ICA) is used to eliminate the interference between signals ,according to the principle of these source signals statistical independence. Secondly, Multiscale Energy Products (MEP) methods estimate the time of arrival. Let signal energies from coarse to finer time scales can be exploited to improve leading edge detection performance. Finally, we use the centroid algorithm to estimate the position from TOA. Compared with other two traditional methods, we demonstrate that the proposed TOA estimation has superior performance, and it is a very good method for interference suppression.

Implementing a Parallel Image Edge Detection Algorithm Based on the Otsu-Canny Operator on the Hadoop Platform.

The Canny operator is widely used to detect edges in images. However, as the size of the image dataset increases, the edge detection performance of the Canny operator decreases and its runtime becomes excessive. To improve the runtime and edge detection performance of the Canny operator, in this paper, we propose a parallel design and implementation for an Otsu-optimized Canny operator using a MapReduce parallel programming model that runs on the Hadoop platform. The Otsu algorithm is used to optimize the Canny operator's dual threshold and improve the edge detection performance, while the MapReduce parallel programming model facilitates parallel processing for the Canny operator to solve the processing speed and communication cost problems that occur when the Canny edge detection algorithm is applied to big data. For the experiments, we constructed datasets of different scales from the Pascal VOC2012 image database. The proposed parallel Otsu-Canny edge detection algorithm performs better than other traditional edge detection algorithms. The parallel approach reduced the running time by approximately 67.2% on a Hadoop cluster architecture consisting of 5 nodes with a dataset of 60,000 images. Overall, our approach system speeds up the system by approximately 3.4 times when processing large-scale datasets, which demonstrates the obvious superiority of our method. The proposed algorithm in this study demonstrates both better edge detection performance and improved time performance.

Video Frames Edge Detection of Red Blood Cells: A Performance Evaluation

The boundary structure (shape and size) and numbers of Red Blood Cells (RBCs) play a significant role in controlling human physiology. Several blood related diseases including sickle hemoglobin, polycythemia, thalassemia, anemia, and leukemia that occur due to the alteration of RBCs structures need precise detection, analysis, and subsequent inhibition. Diverse boundary analyses that have already been developed to detect the frames edge of blood samples remained ineffective. Thus, an automated system needs to be designed for exact detection of the RBCs edges. The edges being the main features of image often provide vital information to separate regions within an object or to detect changes in illumination. Accordingly, edge detection is regarded as an important step in the analysis of images and extraction of valuable information. In this paper, we evaluated the performance of video frames edge detectors and compared it with different gradient based techniques including Robert, Sobel, Prewitt and Canny, Zero-crossing (Laplacian, Gaussian and fuzzy logic) by applying them on video of RBCs. Two criteria such as mean square error (MSE) and peak signal to noise ratio (PSNR)were exploited to achieve good quality of video frames edge detectors for RBCs. Canny edge and fuzzy logic based techniques revealed the optimum performance towards the detection of video frames edge of RBCs.

Edge Detection Based On the Characteristic of Primary Visual Cortex Cells

Aiming at the problem that it is difficult to balance the accuracy of edge detection and anti-noise performance, and referring to the dynamic and static perceptions of the primary visual cortex (V1) cells, a V1 cell model is established to perform edge detection. A spatiotemporal filter is adopted to simulate the receptive field of V1 simple cells, the model V1 cell is obtained after integrating the responses of simple cells by half-wave rectification and normalization, Then the natural image edge is detected by using static perception of V1 cells. The simulation results show that, the V1 model can basically fit the biological data and has the universality of biology. What’s more, compared with other edge detection operators, the proposed model is more effective and has better robustness