The ALPR System : A License Plate Recognition System

Automatic Number Plate Recognition (ANPR) was firstly introduced in 1976 and gained much popularity during the last decade along with the improvement of digital cameras. ANPR is an image processing technology which allows extracting vehicle license plate number from digital camera images. It consists of a camera

The number plates are varied from one and another such as different material use for number plate, blur captured image and the image is skew. Based on the results showed in table 4.6, the different types of material and blur image can be segment using bounding box without any distortions. The number plate used different types of material than actual material. The character and the background of number plate is very shinning and it is the modified number plate where the actual height is 5cm. Some of the number plate fails to recognize due to the luminance condition or problematic background of number plates Bounding box method manage to detect the height character and gives correct output in previous

The picture of the vehicle whose number plate is to be distinguished is caught utilizing the advance camera. Next the number plate is removed by right of the bat changing over rgb image I.e., the caught picture to gray scale image with the specific goal to encourage the plate extraction, and increment the handling speed. Shading picture is produced by advance camera is changed over to dim scale picture utilizing. The essential

ā nd cā n include scā ling, cropping, rotā tions, ā s well ā s non-rigid deformā tions. This ā llows networks which include spā tiā l trā nsformers to not only select regions of ā n imā ge thā t ā re most relevā nt (ā t-tention), but ā lso to trā nsform those regions to ā cā nonicā l, expected pose to simplify recognition in the following lā yers. Notā bly, spā tiā

Face recognition is a technology that is capable of identifying a person from a multimedia object such as a

Features for recognition are enhanced and extracted from the stored bitmap image by using digital image processing. Offline signature recognition is known as Optical Signature Recognition (OCR), because the image of writing is converted into bit pattern by an optically digitizing device such as optical camera or scanner. The recognition is done on this bit pattern data for machine-printed or hand-written text [3]. Recognition of machine printed signatures is also a part of Optical Signature Recognition. In offline, methods are less suitable for man-machine communication because no real time interactivity is present. It is suitable for automatic conversion of paper documents to electric documents, which then may be interpreted by computers. Some applications of the off-line recognition are large-scale data processing such as postal address reading; check sorting, office automation for text entry automatic inspection and identification [11].

developed using Mat lab. The results are shown for the English alphabets and numeric. This

Abstract—Present work targets the detection of humans in images and videos. Our focus is on developing robust feature extraction algorithms that encode image regions as high dimensional feature vectors that support high accuracy human/non-human decisions. To test our feature sets we adopt a relatively simple learning framework that uses linear Support Vector Machines to classify each possible image region as a Human or as a non-Human. This work makes three main contributions. Firstly, we introduce grids of locally normalized Histograms of Oriented Gradients (HOG) as descriptors for people detection in static images. The HOG descriptors are computed over dense and overlapping grids of spatial blocks, with image gradient orientation features extracted at fixed resolution and gathered into a high dimensional feature vector. They are designed to be robust to small changes in image contour locations and directions, and significant changes in image illumination and color. Secondly, the Human detection algorithm has to be executed on modern parallel platforms to achieve the detection goal in real time. Thirdly, alert video monitoring station with an indication. It enhance the Real Time Streaming Protocol (RTSP) based Surveillance cameras purpose.

Convolutional neural systems (CNNs) are suitable for unraveling visual record that depend on hand writing recognition task and characterization [1, 3]. They have an adaptable design which do not need to have complex strategies, for instance, momentum, weight rot, structure dependent learning rates or even finely tuning the engineering[1]. CNNs have additionally accomplished the cutting edge comes about for character acknowledgment on the MNIST informational collection of manually written English digit pictures [2].

1. We gave a brief review on hand engineered feature extraction methods and deep learning procedures.

From the outlined image, the system selects the coordinates whose intensity values as 1. It allocates these values to a new variable, namely, intcoord. From this step, we assume that i and j are the two intcoord values. The next step is to calculate the distance from the origin of both these intcoord values using the following formula.

Abstract- In this paper, we are going to develop real time application on college student for automatic detection and recognition of student during academics, followed by display of personal information of students. This application makes proper use of CCTV camera for real time face detection of students of particular college. The proposed application can be divided into four major steps. In first step, each person in the image is detected. In the second step, a face detection algorithm detects faces of each person. In third step, we use a face recognition algorithm to match the faces of persons in the captured image with the database of students’ faces which also stores personal as well as academic information of each student. In final step, the face of student along with his/her personnel information will be displayed on screen to the user when the image captured by CCTV camera contains any student image of present college. The college administrator as well as faculty members can use this application to identify students and also to distinguish students from outsiders.

Deep architectures are algorithms that apply successive non-linear transformations to data, in an attempt to discover higher and higher level representations of this data [1]. The most common example is an Artificial Neural Network (ANN) with multiple hidden layers. While deep learning is an old concept, learning algorithms on deep architectures were slow and thus impractical [2]. Discovery of new training methods and improved processing power in recent years have allowed for state-of-the-art empirical results in computer vision and speech recognition. Most famously, a group of researchers from Stanford and Google trained a deep network on millions of unlabeled images, which was then able to detect human faces, bodies, and cats [3]. Deep

In Offline method, writing is captured by scanner and final text is formed. The data formed from 40point feature extraction is used to feed artificial neural network to train it and produce more accurate result. Several applications such as document reading, mail sorting, postal address recognition and bank processing require offline handwritten recognition. As a result, offline method is very active area for research for more accurate solutions. In this project we used english alphabet characters to train artificial neural network and test its accuracy of character recognition.

The captured input image has number plate covered by vehicle body, so by this step only number plate area is detected and extracted from whole body of vehicle. The number plate extraction phase influence the accuracy of ANPR system because all further step depend on the accurate extraction of number plate area. The input to this stage is vehicle image and output is a portion of image containing the exact number plate. Number plate can be distinguished by its features. Instead of processing every pixel, the system processes only the pixels that have these features. The features are derived from number plate format and the characters constituting the plate. Number plate color, rectangular shape of number plate boundary, the color change between the plate background and characters on it etc. can be used for detection and extraction of number plate area. The extraction of Indian number plate is difficult as compared to the foreign number plate because in India there is no standard followed for the aspect ratio of plate. This factor makes the detection and extraction of number plate very difficult. The various methods used for number plate extraction are as follows:-