Font Size: a A A

Progressive Transmission Used In Mobile Visual Search

Posted on:2016-08-12Degree:MasterType:Thesis
Country:ChinaCandidate:S Y MiaoFull Text:PDF
GTID:2308330461977060Subject:Software engineering
Abstract/Summary:PDF Full Text Request
Mobile visual search (MVS) is a new class of applications which use the camera phone to initiate search queries about objects in visual proximity to user, which has become a hot topic in both academia and business circles. The reason for this phenomenon is that with the popularization of high-resolution camera embedded mobile devices, mobile phones have evolved into powerful image video processing devices, equipped with color displays, and hardware-accelerated graphics. They are also equipped with location sensors, GPS receivers, and connected to broadband wireless networks allowing fast transmission of information.Many of these applications based on mobile visual search follow a client-server architecture. Due to the bandwidth constrained wireless link, data transmitted process is very slow and costs too much time between the client and the server. And transmission overload turns out to be a bottleneck in most existing mobile visual search applications, leading to low user experience.In order to break this bottleneck, this paper provides a new data transmission strategy as known as progressive transmission based on wavelet used in mobile visual search, which is more suitable for mobile visual search. This strategy can transmit image data in a progressive way, using multi-level wavelet decomposition and reconstruction theorem. And also builds a multi-level Bag-of-Word model on server side, to reconstruct the image and retrieve it.At first, this paper introduces some background and researching meaning of this topic briefly, and makes a detailed introduction of the key technologies involved in this paper, including image descriptor, progressive transmission, bag-of-word model and so on. Then, the system framework of progressive transmission proposed in this paper is illustrated in detail, along with the main idea and the realization of the core algorithm. In the final, extensive experiments have been done on the public Stanford MVS database, comparing to other progressive transmission strategy and demonstrating that the progressive transmission strategy proposed in this paper can achieve higher image retrieval rate with less transmitted image data, reducing transmission latency and enhance use experience.
Keywords/Search Tags:Mobile Visual Search, Progressive Transmission, Wavelet Theorem, Image Descriptor, Bag-of-Word Model
PDF Full Text Request
Related items