Notes on edge detection approaches

Muntarina, Kanija; Shorif, Sumaita Binte; Uddin, Mohammad Shorif

doi:10.1007/s12530-021-09371-8

Cited by 13 publications

(9 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning based methods typically involve training a deep neural network to extract the features important for edge detection on the given dataset of images. These methods have been shown to be highly effective with state-of-the-art performance on many edge detection benchmarks [17]. One example of a recent paper on this topic is "Red-Green-Blue-Dense (RGB-D) Salient Object Detection with Cross-View Generative Adversarial Networks" by [30].…”

Section: Modern Edge Detection Techniquesmentioning

confidence: 99%

“…An image edge detection can be achieved by separating discontinuities in image intensity, colour, or texture [13,14]. Over the years numerous edge detection techniques have been proposed ranging from classical gradient-based methods [2,3] to more recent approaches based on machine learning [15,16] and deep learning [17,18]. Image edge detection is mostly a processing step in myriads of computer vision and image processing tasks such as image segmentation, object recognition and image compression [19].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Image edge detection using pseudo-Boolean polynomials

Chikake,

Goldengorin

2024

Sixteenth International Conference on Machine Vision (ICMV 2023)

View full text Add to dashboard Cite

We introduce a novel approach for image edge detection based on calculating pseudo-Boolean polynomials on image patches whose resulting polynomial degrees determine whether a patch lies over an edge or a blob. In this paper we show that patches covering edge regions within the image result in pseudo-Boolean polynomials of higher degrees compared to patches that cover blob regions. The proposed approach is based on reduction of polynomial degree and equivalence properties of penalty-based pseudo-Boolean polynomials.

show abstract

Section: Modern Edge Detection Techniquesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Image edge detection using pseudo-Boolean polynomials

Chikake,

Goldengorin

2024

Sixteenth International Conference on Machine Vision (ICMV 2023)

View full text Add to dashboard Cite

show abstract

“…The edge detection [18] and simulated phosphene [35] of an image is shown in Figure 2. The edges do not provide a clearer view due to the presence of irrelevant objects and background information.…”

Section: Related Workmentioning

confidence: 99%

“…The existing techniques used for image enhancement primarily focus on capturing the edge [18] and background [19] information. Although the image is enhanced, the results are promising for images having multiple objects [20,21].…”

Section: Introductionmentioning

confidence: 99%

Optimizing Image Enhancement: Feature Engineering for Improved Classification in AI-Assisted Artificial Retinas

Mehmood,

Ko,

Kim

et al. 2024

Sensors

View full text Add to dashboard Cite

Artificial retinas have revolutionized the lives of many blind people by enabling their ability to perceive vision via an implanted chip. Despite significant advancements, there are some limitations that cannot be ignored. Presenting all objects captured in a scene makes their identification difficult. Addressing this limitation is necessary because the artificial retina can utilize a very limited number of pixels to represent vision information. This problem in a multi-object scenario can be mitigated by enhancing images such that only the major objects are considered to be shown in vision. Although simple techniques like edge detection are used, they fall short in representing identifiable objects in complex scenarios, suggesting the idea of integrating primary object edges. To support this idea, the proposed classification model aims at identifying the primary objects based on a suggested set of selective features. The proposed classification model can then be equipped into the artificial retina system for filtering multiple primary objects to enhance vision. The suitability of handling multi-objects enables the system to cope with real-world complex scenarios. The proposed classification model is based on a multi-label deep neural network, specifically designed to leverage from the selective feature set. Initially, the enhanced images proposed in this research are compared with the ones that utilize an edge detection technique for single, dual, and multi-object images. These enhancements are also verified through an intensity profile analysis. Subsequently, the proposed classification model’s performance is evaluated to show the significance of utilizing the suggested features. This includes evaluating the model’s ability to correctly classify the top five, four, three, two, and one object(s), with respective accuracies of up to 84.8%, 85.2%, 86.8%, 91.8%, and 96.4%. Several comparisons such as training/validation loss and accuracies, precision, recall, specificity, and area under a curve indicate reliable results. Based on the overall evaluation of this study, it is concluded that using the suggested set of selective features not only improves the classification model’s performance, but aligns with the specific problem to address the challenge of correctly identifying objects in multi-object scenarios. Therefore, the proposed classification model designed on the basis of selective features is considered to be a very useful tool in supporting the idea of optimizing image enhancement.

show abstract

“…In the images, the Yellow River is widely distributed, has strong morphological features and is obviously different from its surroundings; therefore, the image edge extraction algorithm was used to extract the edges and then vectorize them to determine the variation in the shortest distance between the Yellow River and SHS site in the 1960s and 1970s. The commonly used edge extraction operators are the Sobel, Roberts, Prewitt, Canny, Laplacian of Gaussian (LOG), and Canny of Gaussian (COG) operators [64]. Among these operators, the Sobel, Roberts, Prewitt, and Canny operators are not well balanced between noise and edge extraction accuracy.…”

Section: Edge Information Extraction For the Yellow River At The Nort...mentioning

confidence: 99%

How Can We Understand the Past from Now On? Three-Dimensional Modelling and Landscape Reconstruction of the Shuanghuaishu Site in the Central Plains of China

et al. 2022

View full text Add to dashboard Cite

The Shuanghuaishu (SHS) site in China is one of the 100 most important archaeological discoveries over the past 100 years; its historical heritage can be traced directly back 5300 years. Understanding the early landscape of the site would provide important information about the origin of Chinese civilization. The SHS site is buried and surface traces are difficult to see; therefore, we attempted to reconstruct the early landscape of the site based on a current surface landscape model and environmental archaeological analysis. We created a modern three-dimensional (3D) landscape model of the study area from high spatial resolution unmanned aerial vehicle (UAV) aerial photographs and analysed the distance change between the Yellow River and SHS site in the past 60 years from CORONA and Landsat images. By combining environmental archaeological survey results, archaeological excavation data, relevant papers, and field measurements, we reconstructed the paleotopography of the SHS site during the Yangshao period (7000–5000 aBP). On this basis, 3D natural and human landscapes during the Yangshao period were rebuilt. The results show that (1) Satellite images acquired at different resolutions can provide multiscale spatial information about the site, and high-precision models of current conditions can be quickly generated from UAV aerial photography. (2) From 1960 to 2020, the shortest distance between the SHS site and the Yellow River was approximately 512 m. The location of bedrock on Mang Mountain can be used to infer the early extent of the northern terrace at the site. (3) Environmental archaeology provided information about the palaeoenvironment of the site area. By incorporating spatial information technology and 3D visualization, we can better restore the early landscape of the SHS site. Our work integrates environmental archaeology, field archaeology, and spatial technology, enabling data and modelling support for the visual interpretation of the SHS site.

show abstract

Notes on edge detection approaches

Cited by 13 publications

References 72 publications

Image edge detection using pseudo-Boolean polynomials

Image edge detection using pseudo-Boolean polynomials

Optimizing Image Enhancement: Feature Engineering for Improved Classification in AI-Assisted Artificial Retinas

How Can We Understand the Past from Now On? Three-Dimensional Modelling and Landscape Reconstruction of the Shuanghuaishu Site in the Central Plains of China

Contact Info

Product

Resources

About