Candidate Feature Points Research Articles

The Flatness Assessment Algorithm Based on the Linear Filter

According to modern digital flatness assessment attributes, it is required to eliminate interference points by the most simple and fast algorithm, reducing scanning time of data and then optimizing the flatness assessment algorithm. The flatness algorithm by linear filter is suggested to seek out the candidate feature points of flatness, on the smallest inclusion surface, with no more than two high (or low) points on one line, no more than three high (or low) points on one flat surface, and we therefore propose the corresponding searching methods. This model is easy to implement in a computer, and the algorithm is simple, decreasing the number of searches, and also guarantee to be consistent with the new generation of GPS precise interpretation of the principle of minimum zone.

Linear Scale and Rotation Invariant Matching

Matching visual patterns that appear scaled, rotated, and deformed with respect to each other is a challenging problem. We propose a linear formulation that simultaneously matches feature points and estimates global geometrical transformation in a constrained linear space. The linear scheme enables search space reduction based on the lower convex hull property so that the problem size is largely decoupled from the original hard combinatorial problem. Our method therefore can be used to solve large scale problems that involve a very large number of candidate feature points. Without using prepruning in the search, this method is more robust in dealing with weak features and clutter. We apply the proposed method to action detection and image matching. Our results on a variety of images and videos demonstrate that our method is accurate, efficient, and robust.

Interest operators for feature‐based matching in close range photogrammetry

AbstractAlthough automated object surface reconstruction via feature‐based matching is commonly employed in both stereoscopic close range and topographic photogrammetry, it has rarely been used in conjunction with multi‐image, convergent photogrammetric networks. A prerequisite for feature‐based matching of distinct points is the application of interest operators to provide the dense arrays of candidate feature points within the images forming the network. This paper describes an evaluation of three interest operators, the Förstner, SUSAN and FAST algorithms, the aim being to assess which is optimal for feature‐based matching in convergent close range photogrammetry. Following a brief review of the development of interest operators, a description of the three operators is provided, with the recently developed FAST operator being discussed in more detail. The merits of image pre‐processing via the Wallis filter are also outlined, after which the performance of the interest operators is experimentally assessed within an eight‐image network on the basis of accuracy of interest‐point localisation, detection rate and speed. The findings of the evaluation are that, of the Förstner, SUSAN and FAST operators, the FAST, which has not been employed to a significant extent in photogrammetry to date, is optimal for feature‐based matching in multi‐image close range networks.

Detection of closed sharp edges in point clouds using normal estimation and graph theory

The reconstruction of a surface model from a point cloud is an important task in the reverse engineering of industrial parts. We aim at constructing a curve network on the point cloud that will define the border of the various surface patches. In this paper, we present an algorithm to extract closed sharp feature lines, which is necessary to create such a closed curve network. We use a first order segmentation to extract candidate feature points and process them as a graph to recover the sharp feature lines. To this end, a minimum spanning tree is constructed and afterwards a reconnection procedure closes the lines. The algorithm is fast and gives good results for real-world point sets from industrial applications.

Extension of Hidden Markov Models for Multiple Candidates and Its Application to Gesture Recognition

We propose a modified Hidden Markov Model (HMM) with a view to improve gesture recognition using a moving camera. The conventional HMM is formulated so as to deal with only one feature candidate per frame. However, for a mobile robot, the background and the lighting conditions are always changing, and the feature extraction problem becomes difficult. It is almost impossible to extract a reliable feature vector under such conditions. In this paper, we define a new gesture recognition framework in which multiple candidates of feature vectors are generated with confidence measures and the HMM is extended to deal with these multiple feature vectors. Experimental results comparing the proposed system with feature vectors based on DCT and the method of selecting only one candidate feature point verifies the effectiveness of the proposed technique.