Umeå University's logo

In this paper we propose a multi-scale edge detection algorithm based on proportional scale summing. Our analysis shows that proportional scale summing successfully improves edge detection rate by applying independent thresholds on multi-scale gradient images. The proposed method improves edge detection and localization by summing gradient images with a proportional parameter cⁿ (c < 1); which ensures that the detected edges are as close as possible to the fine scale. We employ non-maxima suppression and thinning step similar to Canny edge detection framework on the summed gradient images. The proposed method can detect edges successfully and experimental results show that it leads to better edge detection performance than Canny edge detector and scale multiplication edge detector.

In this paper, a novel edge detection method that computes image gradient using the concept of Center of Mass (COM) is presented. The algorithm runs with a constant number of operations per pixel independently from its scale by using integral image. Compared with the conventional convolutional edge detector such as Sobel edge detector, the proposed method performs faster when region size is larger than 9×9. The proposed method can be used as framework for multi-scale edge detectors when the goal is to achieve fast performance. Experimental results show that edge detection by COM is competent with Canny edge detection.

In edge detection algorithms, there is a common redundancy problem, especially when the gradient direction is close to -135°, -45°, 45°, and 135°. Double edge effect appears on the edges around these directions. This is caused by the discrete calculation of non-maximum suppression. Many algorithms use edge points as feature for further task such as line extraction, curve detection, matching and recognition. Redundancy is a very important factor of algorithm speed and accuracy. We find that most edge detection algorithms have redundancy of 50% in the worst case and 0% in the best case depending on the edge direction distribution. The common redundancy rate on natural images is approximately between 15% and 20%. Based on Canny’s framework, we propose a restriction in the hysteresis step. Our experiment shows that proposed restricted hysteresis reduce the redundancy successfully.

Effective and efficient generation of keypoints from images is the first step of many computer vision applications, such as object matching. The last decade presented us with an arms race toward faster and more robust keypoint detection, feature description and matching. This resulted in several new algorithms, for example Scale Invariant Features Transform (SIFT), Speed-up Robust Feature (SURF), Oriented FAST and Rotated BRIEF (ORB) and Binary Robust Invariant Scalable Keypoints (BRISK). The keypoint detection has been improved using various techniques in most of these algorithms. However, in the search for faster computing, the accuracy of the algorithms is decreasing. In this paper, we present SICK (Scale-Invariant Corner Keypoints), which is a novel method for fast keypoint detection. Our experiment results show that SICK is faster to compute and more robust than recent state-of-the-art methods. 

The first step of efficient edge detection is to use a filter to detect intensity change. The filter size is a parameter which affects the edge detection result. A filter of large size is less sensitive to noise while a filter of small size is more accurate when locating edges. This gives the user a choice of choosing the proper filter size depending on the situation. A more stable edge detection approach is multi-scale edge detection, which detects gradients using several filter sizes.  The time consumption of a conventional edge filter is usually  or , where w is the width of the filter. Therefore, using filters of large size or multi-scale filters is not very efficient. We propose an efficient edge detection method with  time consumption. It uses the center of mass concept and utilizes the power of integral images to achieve this efficiency. The results of our experiments show that the proposed edge detector is very stable and we also propose a simplified multi-scale edge detection scheme which can be used in practical operations.  

Curves and lines are geometrical, abstract features of an image. Whereas interest points are more limited, curves and lines provide much more information of the image structure. However, the research done in curve and line detection is very fragmented. The concept of scale space is not yet fused very well into curve and line detection. Keypoint (e.g. SIFT, SURF, ORB) is a successful concept which represent features (e.g. blob, corner etc.) in scale space. Stimulated by the keypoint concept, a method which extracts distinctive curves (DICU) in scale space, including lines as a special form of curve features is proposed. A curve feature can be represented by three keypoints (two end points, and one middle point). A good way to test the quality of detected curves is to analyse the repeatability under various image transformations. DICU using the standard Oxford benchmark is evaluated. The overlap error is calculated by averaging the overlap error of three keypoints on the curve. Experiment results show that DICU achieves good repeatability comparing with other state-of-the-art methods. To match curve features, a relatively uncomplicated way is to combine local descriptors of three keypoints on each curve.

This paper aims to broaden the scope of shape related features including edges, corners, lines and curves: 1) Edges, corners, lines, curves are all shape related features. In the past, the detection of each type of feature is usually solved independently under certain hypotheses. Our proposed distinctive curve detection method (DICU) solves the detection of all these type of features together. 2) Compared to the development in scale-invariant interest point detectors which have adopted more objective robustness measures using repeatability score, the research in line and curve features is still limited to “true/false positive” measures. DICU detection utilizes the scale-space concept and proves that curve features can be as robust as scale-invariant interest points. DICU has three advantages: 1) DICU outputs multi-type features which can benefit future computer vision applications. At the same time, the computational efficiency is unaffected, after detecting edges, only 5% additional computation is needed to detect corners, lines, and curves. 2) It is robust under various image perturbations and transformations and outperforms state-of-the-art interest point detectors and line detectors. At the same time, all types of detected features are robust. 3) Curve features contains more geometric information than points. Our curve matching test shows that curve matching can outperform interest point matching. 

This paper presents a new methodology that matches image geometry using a curve chain. A curve chain is defined as a 1-dimensional arrangement of curves. The idea is to match images without using local descriptors and apply this concept into applications. This paper have two contributions. First, we present a novel curve feature which is scale & rotation – invariant. Secondly, we present an efficient scale & rotational-invariant matching method which matches curve chains in the scene. The efficacy is benefited by three factors. Firstly, matching a 1-dimensional curve chain can achieve quadratic operations when dynamic programming is used.  Secondly, curves are salient features that naturally reduce the dimensionality compared with scanning all possible locations. Thirdly, curves provide stable relational cues between neighbouring curves. Such stable relational cues reduce the computation to linear operations by avoiding searching all combinations of curves in dynamic programming. The advantages of the method has good potential to benefit application including point correspondence matching, object detection, etc.  In point correspondence experiments our method yields a good total matching score on various image transformations. At the same time, the proposed method shows good potential of matching non-rigid object such as faces with scale & rotation invariance.