Skin detection

Skin Detection

Skin color has proven to be a useful and robust cue for face detection, localization and tracking. Image content filtering, content aware video compression and image color balancing applications can also benefit from automatic detection of skin in images. Numerous techniques for skin color modelling and recognition have been proposed during several past years.Color usually presents a strong and intuitive cue in images with complex scenes. 

Automatic target scoring system in video using machine vision

As a popular form of sporting, shooting training is attached great importance not only in athletics but in military as well. Traditionally, the shooting training process, usually including target setting and examining and the recording of shooting results,was performed manually. in this project an automatic shooting scoring system is implemented. this system can help both trainers and athletes do more quality than quantity work.

C3D (chromium Direction Densitr Detector)

Several image processing techniques have been applied in a mine scale exploration for chromite deposits. Traditional exploration methods are based on geophysical methods, which suffer several shortcomings, including lack of sufficient geophysical contrast?. An optic-geometric image processing  program is developed  for extracting structural properties of  chromite minerals in polished sections. The estimated and calculated geometric attributes and parameters based on brightness and morphometric properties of minerals in microscopic scale are stored in a data base.  Computationally, the distinction between blind mineralization and false ore mineralization is possible, without exploration drilling, by this method. The methodology developed in this research, has been validated by testing it on various real world scenarios. It includes the structures of ore fields and chromite deposits in large and mine scale. The end result of this study gives promises for chromites exploration in mine scale using an algorithmic image processing technique. The advantage of the proposed method is that a quantitative method is replaced by a qualitative one. This can lead to make optimal managerial and economical decisions.

Car recognition using SIFT and Color Histogram

Car recognition using SIFT and Color Histogram

Scale invariant feature transform (SIFT) method is invariant to scale, rotation and also is partially invariant to illumination differences and noise which makes it well suited for object recognition. However, employing the SIFT features is time consuming while using a large dataset, which results from complexities of SIFT descriptor matching procedure. The following paper proposes a hierarchical method to recognize a car among 600 random samples from 40 different cars from video streams, based on colour histogram and SIFT features. The proposed method can properly recognize 85% of the cars in the database and decreases the process time up to 47% compared with the time that is needed to recognize only with SIFT. Furthermore, using sequence of frames lets the system work properly in different scenes.

Multi-camera People Tracking

multi-camera people Tracking

Object tracking is the task of approximating the trajectory of an object as it moves in a certain area. In most cases, covering the complete surveillance area is not feasible by a single camera due to limitations of cameras field of view (FOV) and the structure of scene. Therefore, surveilling a wide area requires a system with the ability to track objects across multiple cameras. Moreover, it is not feasible to completely cover wide areas, such as shopping centers, by an overlapping camera network due to the structure of the environment, and computational constraints. Thus, in realistic scenarios, the system should be able to handle multiple cameras with non-overlapping fields of view. Each camera observes a disjoint part of the environment that Tracking in such situation requires camera-to-camera correspondence of observations (I.e. re-identification of an object) when an object leaves one scene and appears at another scene later. Tracking people as they move through a camera network with non-overlapping field of view is a challenging issue.

In this project ( My M.Sc thesis) a multi-camera tracking system is proposed, where a number of stationary cameras are installed at an indoor environment and have non-overlapping fields of view. Tracking in such situation requires re-identification of objects when they leaves one field of view and appears at another scene later. Tracking in such systems is a challenging issue due to the changes of their appearance among the cameras. Appearance of a pedestrian is represented by its color histogram that is the consequence of illumination changes, parameters of camera, viewing angle and materials of the clothing. In addition in real world multi camera systems, the observations of people are often occurred arbitrarily in a widely separated time and position, and common prediction techniques such as Kalman filter cannot be used to establish correspondence between observations from different cameras.

Free WordPress Themes - Download High-quality Templates