![]() It is necessary to clarify that it exploits the Computer Vision Toolbox, and as a consequence, it is required for proper implementation. It is a simple and easy-to-use command-line function written in MATLAB R2020a. VISION also gives the possibility to adopt a region of interest (ROI) – on which the stabilisation analysis is carried out – and visualise stabilisation performances in real-time. Afterwards, the strongest and uniformly distributed features are kept and used to stabilise. This choice can be beneficial by keeping the best algorithms capabilities but at the expenses of higher computational loads. Seven feature detection algorithms are implemented, and in order to exploit as maximum as possible the different and own algorithm capabilities, the stabilisation analysis can be carried out with two of them chosen simultaneously. The user’s involvement is, therefore, little required being a good option for both experienced and non-experienced ones. Hence, stabilisation can be carried out with and without ground control points (GCPs), giving more flexibility to the user and case studies under natural conditions. The latter detects features automatically on a frame-by-frame basis and match them to stabilise the video. VISION aims to overcome the aforementioned issues by providing an easy-to-use and open-code MATLAB command-line function following an automatic feature selection approach. Features are matched using the Kanade–Lucas–Tomasi (KLT) tracking algorithm with five pyramid levels, and frames are stabilised using a similarity transform. Afterwards, the field of view is divided into four quadrants, and the strongest 10% of detected features – using the minimum eigenvalue algorithm – are kept. This zone can be, for instance, the area where the water flows or other regions with clear movements. The stabilisation is not fully automated, requiring user involvement by drawing a zone to be excluded due to evident motion (mask). The stabilisation process is performed taking as reference the first frame, which defines the coordinate system, and subsequent frames are referred to it. Different modules are implemented, giving endless possibilities to carry out image velocimetry and pre-processing tasks. Flexibility within the analysis is provided by the fact that surface flow velocities can be determined using either mobile camera platforms or fixed monitoring stations. ] is an easy-to-use software package for sensing flow velocities and discharge in rivers. VISION is an easy-to-use software that may support research operating in image processing, but it can also be adopted for educational purposes. In particular, the stabilisation impact was quantified in terms of velocity errors with respect to field measurements obtaining a significant error reduction of velocities. One case study was deemed to illustrate VISION stabilisation capabilities on an image velocimetry experiment. ![]() It includes a number of options that can be set depending on the user’s needs and intended application: 1) selection of different feature detection algorithms (seven to be selected with the flexibility to choose two simultaneously), 2) definition of the percentual value of the strongest features detected to be considered for stabilisation, 3) geometric transformation type, 4) definition of a region of interest on which the analysis can be performed, and 5) visualisation in real-time of stabilised frames. ![]() It can be applied for any use, but it has been developed mainly for image velocimetry applications in rivers. VISION is open-source software written in MATLAB for video stabilisation using automatic features detection.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |