Browsing by Subject "Computer vision"
Now showing items 1-2 of 2
-
(2020)Omnidirectional microscopy (OM) is an emerging technology capable of enhancing the threedimensional (3D) microscopy widely applied in life sciences. In OM, precise position and orientation control are required for the sample. However, the current OM technology relies on destructive, mechanical methods to hold the samples, such as embedding samples in gel or attaching them to a needle to permit orientation control. A non-contacting alternative is the levitation of the sample. But, until now, the levitation methods have lacked orientation control. I enable omnidirectional access to the sample by introducing a method for acoustic levitation that provides precise orientation control. Such control around three axes of rotation permits imaging of the sample from any direction with a fixed camera and subsequent 3D shape reconstruction. The control of non-spherical particles is achieved using an asymmetric acoustic field created with a phase-controlled transducer array. The technology allows 3D imaging of delicate samples and their study in a time-lapse manner. I foresee that the described method is not only limited to microscopy and optical imaging, but is also compatible with automated sample handling, light-sheet microscopy, wall-less chemistry, and noncontacting tomography. I demonstrate the method by performing a surface reconstruction of three test samples and a biological sample. In addition, a simulation study and the levitation of test samples were used to characterize the levitation technique's performance. Both the shape reconstruction and orientation recovery were done by a computer vision based approach where the different images are stitched together. The results show the rotation stability and the wide angle range of the method.
-
(Helsingin yliopistoHelsingfors universitetUniversity of Helsinki, 2011)Modern smart phones often come with a significant amount of computational power and an integrated digital camera making them an ideal platform for intelligents assistants. This work is restricted to retail environments, where users could be provided with for example navigational instructions to desired products or information about special offers within their close proximity. This kind of applications usually require information about the user's current location in the domain environment, which in our case corresponds to a retail store. We propose a vision based positioning approach that recognizes products the user's mobile phone's camera is currently pointing at. The products are related to locations within the store, which enables us to locate the user by pointing the mobile phone's camera to a group of products. The first step of our method is to extract meaningful features from digital images. We use the Scale- Invariant Feature Transform SIFT algorithm, which extracts features that are highly distinctive in the sense that they can be correctly matched against a large database of features from many images. We collect a comprehensive set of images from all meaningful locations within our domain and extract the SIFT features from each of these images. As the SIFT features are of high dimensionality and thus comparing individual features is infeasible, we apply the Bags of Keypoints method which creates a generic representation, visual category, from all features extracted from images taken from a specific location. A category for an unseen image can be deduced by extracting the corresponding SIFT features and by choosing the category that best fits the extracted features. We have applied the proposed method within a Finnish supermarket. We consider grocery shelves as categories which is a sufficient level of accuracy to help users navigate or to provide useful information about nearby products. We achieve a 40% accuracy which is quite low for commercial applications while significantly outperforming the random guess baseline. Our results suggest that the accuracy of the classification could be increased with a deeper analysis on the domain and by combining existing positioning methods with ours.
Now showing items 1-2 of 2