IJRIT International Journal of Research in Information Technology, Volume 2, Issue 6, June 2014, Pg: 114-118

International Journal of Research in Information Technology (IJRIT)

www.ijrit.com

ISSN 2001-5569

Door Detection and Door State Identification Jaymit B. Pandya and Prem D. Balani Nikita R. Patel Student G.H.Patel college of Engineering and Technology V.V.Nagar, Anand Gujarat-India

Assistant Professor Information Technology Department G.H.Patel college of Engineering and Technology V.V.Nagar, Anand Gujarat-India

[email protected]

[email protected], [email protected]

Abstract – This Article present a method for door detection by considering main features of door as feature extraction. Region of interest (ROI) is considered for extraction of features. The aim of the work presented here is to develop a method door state identification by considering edges and angle. Experimental results show that our new method of door detection is efficient for differentiate with window, other corridors and identify state of a door. Keywords-component – Door detection; State identification; Computer vision;CHT

I. INTRODUCTION

A series of approaches have been proposed by different research group in the field of computer vision to the problem of recognizing doors by visual characteristics such as color, texture, features. Indoor semi-structured environments are full of corridors that connect different offices and laboratories where doors give access to many of those locations that are defined as goals for the robot as well other fields. Door represents the entrance and exit points of rooms. Worldwide there are about 285 million visually impaired persons, out of them 39 million are blind and the others have low vision. The World Health Organization (WHO) estimates that there were 161 million visually impaired people around the world in 2002, about 2.6% of the total population. So, among this figure, 124 million (about 2%) had low vision and 37 million (about 0.6%) were blind. In current era for robot self navigation and localization, for blind person door detection is challenging application as well in gaming environment door plays important role. So, Image processing is providing the direction for solving this challenging task. Adaboost is one of the best approaches considered for door identification. This algorithm captures all features of door like base, gap, handle, frame, color but using this it’s not possible to capture all features in a single pc camera. Different edge detection techniques like Canny, Robert, Sobel and Prewitt are used for detection and canny gives best result as compare to other methods and able to differentiate door from windows and other corridors. Corner detection is used for detecting horizontal and vertical lines along with its corners but not able to give presence of door by differentiating with window. The door detection problem becomes very hard in real situation. Almost all systems designed to assist them are quite complex and expensive, but most application do not have advanced technical assistance and they are rather poor. Finding door status is become needful requirement. So algorithm which efficiently detect door and detect its status along with these requirements should required. This thesis focuses on efficient door detection and status identification method. II. RELATED RESEARCH It is still an open problem to detect door in an unfamiliar environment because doors seems like windows and other corridors as well door constitutes different size, material, color so, it would be better to make a more theoretical analysis on the door detection approach. Some improvement should be made to the approach to tackle some complicated issues automatically. For

Nikita R. Patel,IJRIT

114

IJRIT International Journal of Research in Information Technology, Volume 2, Issue 6, June 2014, Pg: 114-118

example, by considering common features of each and every door like knob one can detect any kind of door by differing from window and other corridors but it is still open issue to detect any kind of knob because knob can be horizontal, vertical, and circular so shape detection is required. In recent years there has been much work done on edge detection using image processing approach that are described by author in [1]. We have described some work done on door detection using edge detection. In [2] Jens Hensler has proposed Adaboost algorithm for door detection using key features of door. So their proposed technique ensures presence of door. In [3] Xiaodong Yang has proposed door detection method by combining edges and corners features that utilizes the general and stable features of doors - edges and corners. Furthermore it presents a general geometric model to characterize the door shape by combining edge and corner features without a training process. In [4] E. Jauregi has proposed the new approach of door Handle Identification the segmentation with the Hough transform and statistical measures is used by 3-stage process: Shape Detection, Colour Segmentation and Statical Validation. In [5] authors have proposed a method in which extracts a region of interest (ROI) from image that with high probability that contains the handle. The method for extracting the ROI are compared, Circle Hough Transform (CHT) and combined with three descriptor extraction methods: SIFT, SURF and USURF. In [6] author has proposed a method that utilized the general and stable features, edges and corners to implement door detection. After detecting the door, to indicate the door’s position to the blind user. Classify the relative position as Left, Front, and Right. In [8] an author focuses mainly on the image segmentation using edge operators. The interaction between image segmentation and object recognition in the framework of the Sobel, Prewitt, Roberts, Canny. Canny operator performed better than Sobel, Prewitt, Roberts and LoG. The EM algorithm produced stable segmentation effect on different types of images. OTSU showed good and stable segmentation effect In [7] Rafiq sekkal has proposed the method that estimates the wall/floor boundary. To this aim, lines are extracted and further cross-checked with temporally consistent vanishing point .Then rectangular shapes that respect a door-like height/width ratio are extracted within wall planes. So that was the different work carried in door detection till now. In next section we will see how our technique will detect door and identify state of door. III. PROPOSED WORK

In our work we have considered two main features of door key hole and handle for door detection which differentiate door from window and other corridors. For door state identification angle and line detection is considered.

A. Door Detection An algorithm for proposed method of door detection is implemented as below: Acquire the image from image database. Perform further preprocessing on the image.Apply thresholding on preprocessed image and apply ROI. Then by Performing image segmentation recognize object.Lastely extract features of door which uniquely detect door. B. Door State Identification An algorithm for state identification is based on below given condition: 1. Condition 1: Does the end points of the first and last long vertical lines belong to the baseboard line? 2. Condition 2: Is the angle determines by the second pair of correlation window is 0 or positive in the clockwise or anticlockwise direction in relation to the baseboard line? 3. Condition 3: Does the third pair of correlation window is applied to identify a non-vertical line? If yes, then what is the angle with line? IV. IMPLEMENTATION RESULTS Following are the results of different door detection method by considering image as a input. (A) Door Detection

Nikita R. Patel,IJRIT

115

IJRIT International Journal of Research in Information Technology, Volume 2, Issue 6, June 2014, Pg: 114-118

1) Canny Edge detection

(a)

(b)

Figure 1.(a) Original image (b) Result using Canny 2) Corner Extraction

(a) (b) Figure 2.(a) Original image (b) Result of corner extraction 3) Features Extraction

Nikita R. Patel,IJRIT

(a)

(b)

(c)

(d)

116

IJRIT International Journal of Research in Information Technology, Volume 2, Issue 6, June 2014, Pg: 114-118

(e) Figure 3.(a) Original image (b) Gray scale (c) Binary image(d) Centroid of an image (e) Feature Extraction (B) Door State Identification

(a)

(c)

(b)

(d)

Figure 4.(a), (b) Line is on baseboard (c),(d) Line is at angle Results in figure 1(a), 1(b) of canny edge detection techniques shows it requires performing smoothing on an image to extract feature as well door boundary which improves the result. Results in figure 2(a), 2(b) of corner extractions shows it doesn’t gives false positive rate of door detection by extraction of each and every corners, it may be window or other corridors so it requires further processing and other features for door detection. Results in figure 3(a), 3(b), 3(c), 3(d), 3(e) of feature extraction shows the original image is first, converted into gray scale for edge detection second, the gray scale image is converted into binary image to recognize shape or features. Then, in the gray scale image centroid detection is preformed for detecting knob (By assuming knob is in centre of door) and last, HSV conversion is performed to do process on knob and key-hole based on color segmentation. Results in figure 4(a), 4(b), 4(c), 4(d) of state identification shows the vertical lines are on baseboard and doesn’t making any angle then door state is close and the lines are on baseboard and making any angle in positive clock-wise or anti-clockwise then door state is open. V. CONCLUSION & FUTURE WORK In this paper we have described how we can use feature extraction method and improve the efficiency of the door detection.From experiments and results,our proposed method for door detection conclude that we can detect door by differentiating it with windows and other corridors. The new method successfully identifies door states by distinguishing between open and close doors without any previous knowledge of presented environment.

Nikita R. Patel,IJRIT

117

IJRIT International Journal of Research in Information Technology, Volume 2, Issue 6, June 2014, Pg: 114-118

The future work can includes to identify other door types such as the elevator doors, the stairways etc by modifying the algorithm. Also porting the software to c++ and perform the identification by motion analysis and prediction. ACKNOWLEDGEMENT My Sincere thanks to my guide Prof. Jaymit B. Pandya and Co-guide Prof. Prem D. Balani for providing me an opportunity to do my research work. I express my thanks to my Institution namely G. H. Patel College of Engineering and Technology for providing me with a good environment and facilities like Internet, books, computers and all that as my source to complete this research work. My heart-felt thanks to my family, friends and colleagues who have helped me for the completion of this work. REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]

Using the Canny Edge Detector for Feature Extraction and Enhancement of Remote Sensing Images, Mohamed Ali, David Clausi- 2001(IEEE) Adaboost Based Door Detection For Mobile Robot,Jens Hensler, Michael Blaich, Oliver Bittel, 2010, International Conference on Agents and Artificial Intelligence(ICAART) Robust Door Detection in Unfamiliar Environments by Combining Edge and Corner Features, Xiaodong Yang ,Yingli Tian, 2010, IEEE Door Handle Identification: A Three Stage Approach ,E. Jauregi, J.M.Mart´ınezOtzeta, B. Sierra, E. Lazkano, 2007, Computer Science and Artificial Intelligence Department University of the Basque Country (IFAC) Visual Approaches For Handle Recognition,E. Jaurgi, E.Lazkano, J.M.Mart’inez otzeta, B. Sierra, Robotics and autonomous system group University of Basque country, Donostia-2008(springer) Context-Based Indoor Object Detection As An Aid To Blind Person Accessing Unfamiliar Environments, Xiaodong Yang and Yingli Tian , Chucai Yi, Aries Arditi, ACM,2010 Simple Monocular Door Detection And Tracking ,Rafiq sekkal, Francois Pasteau, Marie Babel, Baptiste Brun, Ivan Leplumey, 2013-IEEE Realtime local navigation for the blind: detection of lateral doors and sound interface, M. Moreno, S.Shahrabadi, J.Jose, J.M.H.du Buf , J.M.Frodrigues, 2012-Elsevier Segmentation And Object Recognition Using Edge Detection Techniques, Y.Ramadevi, T.Sridevi, B.Poornima, B.Kalyani, 2010-IJCSIT Circular Object Detection Using A Modified Hough Transform, Marcin Smereka, Igancy Duleba- 2008(AMCS) A Self Organizing Neural Scheme For Door Detection In Different Environments F.Mahmood, F.Kuwar- International Journal OF Computer Applications, 2012 Laser-based Perception for Door and Handle Identification , Radu Bogdan Rusu,Wim Meeusseny, Sachin Chittay, Michael Beetz, 2009-ICAR

Nikita R. Patel,IJRIT

118

Door Detection and Door State Identification - International Journal of ...

goals for the robot as well other fields. Door represents the entrance and exit points of rooms. Worldwide there are about 285 million visually impaired persons, ...

133KB Sizes 0 Downloads 220 Views

Recommend Documents

Door Detection and Door State Identification - International Journal of ...
for robot self navigation and localization, for blind person door detection is challenging application as well in gaming environment door plays important role. .... [8] Realtime local navigation for the blind: detection of lateral doors and sound int

Cheap Broadlink S1C Door Sensor 433Mhz Connect Wireless Door ...
Cheap Broadlink S1C Door Sensor 433Mhz Connect Wir ... ecurity System Free Shipping & Wholesale Price.pdf. Cheap Broadlink S1C Door Sensor 433Mhz ...

pdf-1498\-the-apocalypse-door-the-apocalypse-door-by-macdonald ...
... apps below to open or edit this item. pdf-1498\-the-apocalypse-door-the-apocalypse-door-b ... thor-dec-08-2009-paperback-by-james-d-macdonald.pdf.

The Living Door
The way may be rocky and treacherous, but I will not fear; my Shepherd is with me, guiding me with His rod and staff. Thank You for the confidence and security I.

The Living Door
But in John 10, Jesus, the Living Door, opened the way for anyone with an open heart to enter the fold of God. Additionally, Jesus, the. Good Shepherd, gave His ...

next door big.pdf
There was a problem previewing this document. Retrying... Download. Connect more apps... Try one of the apps below to open or edit this item. next door big.pdf.

Jack shaft garage door operator
Apr 12, 2000 - Designer's Notebook, Six Ways to Reduce Costs with Tran ..... engages the pivotable arm. 9. A jack shaft garage door operator for opening and.

Garage Door Cable.pdf
some ideas you might want to try for increasing garage security. Keep in mind, though, that this. only applies to garages and garage door openers that are like ...

Garage Door Cable.pdf
Phone:​ ​(587)​ ​316-5600. Website:​ ​https://garagepros.ca/. Google​ ​Site: https://sites.google.com/site/garagedoorrepairscalgary/. Google​ ​Folder​ ...

Garage Door Cable.pdf
Pinterest https://www.pinterest.com/garageprosca/. Page 3 of 4. Garage Door Cable.pdf. Garage Door Cable.pdf. Open. Extract. Open with. Sign In. Main menu.

OO_NP7_Goed werk door provocatieve professionals_Brom+ ...
OO_NP7_Goed werk door provocatieve professionals_Brom+Bruining.pdf. OO_NP7_Goed werk door provocatieve professionals_Brom+Bruining.pdf. Open.

Garage Door Remote.pdf
Page 3 of 5. Garage Door Remote.pdf. Garage Door Remote.pdf. Open. Extract. Open with. Sign In. Main menu. Displaying Garage Door Remote.pdf. Page 1 of ...

Stage Door Program.pdf
There was a problem previewing this document. Retrying... Download. Connect more apps... Try one of the apps below to open or edit this item. Stage Door ...

Overhead Door Company.pdf
Sign in. Loading… Whoops! There was a problem loading more pages. Retrying... Whoops! There was a problem previewing this document. Retrying.

Garage Door Remote.pdf
Additionally take into consideration whether you desire other means of access– maybe. electronic keypads where you tap a code into active the opening and ...

Jack shaft garage door operator
Apr 12, 2000 - Designer's Notebook, Six Ways to Reduce Costs with Tran ..... engages the pivotable arm. 9. A jack shaft garage door operator for opening and.

Measure Door Window.pdf
THEN: Round up your measurement to the next whole and even number. Ex. If your ... After you round up. and add the ... Displaying Measure Door Window.pdf.