Abstract–Active-pixel CMOS image sensors have many
attractive features, such as low power consumption, integrated
on peripheral circuits, and nondestructive column-parallel
readouts. Integrated on signal-processing circuits, they provide
the advantages of high-speed parallel operation and lower power
consumption, features which meet the special requirements of
CMOS image sensors. A number of integrated signal processing
circuit sensors have been developed and adopted not only for
video but also for machine-vision and security applications as
well. However, these approaches almost always involve lowering
the pixel density and increasing the chip size to accommodate the
added functions requiring large circuits. Hence proposed system
is designed to detect a motion on the basis of difference between
current and previous frame. While finding the difference system
will indicate the location or the position of the motions. System
will also capable to detect motion in dark by reducing the
infrared filtering from CMOS sensor. The system will then give
alert to user through light operating or buzzing the sound.
Keywords’ CMOS image sensors.
I. INTRODUCTION
In market-driven applications such as surveillance,
automotive, and machine vision, there is an increased demand
for imaging systems with real-time processing capabilities. In
some cases, these specific requirements are quite hard to be
fulfilled through a conventional approach, consisting of a
standard charge-coupled device or complementary metal’
oxide’semiconductor (CMOS) camera linked to a digital
signal-processing platform. These systems are typically based
on general purpose architectures, performing real-time image
processing. Although their high computational power and high
flexibility are satisfactory for many applications, there are
some low-level images processing tasks that can be efficiently
executed using ad-hoc image processing capabilities
embedded directly in the imager. Thanks to the great
advantages of CMOS sub micrometer technology, allowing
even smaller device feature size, some recent CMOS image
sensors with integrated signal processing have been
developed, following two main approaches: pixel-level and
array-level processing.
II. PROPOSED IMPLEMENTATION
Camera
Frame Buffer
Live Frame
Comparator
Hardware Control Unit
Central Processing Unit
Figure 1.0 System Architecture
Figure 1.0 describes the proposed system architecture
where the CMOS sensor will be used to capture the frame at
maximum possible speed. These frames are then transferred to
central processor as array of data which will always get store
at buffer of exact size of captured frame. This buffer will be
utilized by the processing unit to find out any changes in
current captured view with respect to buffer image. If there are
any dissimilarity between buffered image and current captured
image more that the defined threshold values then the system
will generate the alert signals [5]. Defining the threshold value
is compulsory CMOS sensor result are environmental changes
dependent so system will never get exact result even after the
camera and the scene is steady. Design a CMOS module
interfacing with controller board system. Develop a software
program to capture a camera view and detect motion in
frames. Build microcontroller based hardware in order to ON
the emergency alert system.
III. RESEARCH METHODOLOGY TO BE EMPLOYED
Proposed system is mainly divided in two following modules.
First is video capturing to get the video frames, next is image
processing to get the images from frame, and next is to get the
pixel information from the image, the detection of color from
pixel and at last controlling the hardware.
III.1 IMAGE ACQUISITION
A digital image is produced by one or several image
sensors, which, besides various types of light-sensitive
cameras, include range sensors, tomography devices, radar,
ultra-sonic cameras, etc. Depending on the type of sensor, the
resulting image data is an ordinary 2D image, a 3D volume, or
an image sequence. The pixel values typically correspond to
light intensity in one or several spectral bands (gray images or
colour images), but can also be related to various physical
measures, such as depth, absorption or reflectance of sonic or
electromagnetic waves, or nuclear magnetic resonance.
III.2 PRE-PROCESSING
Before a computer vision method can be applied to image
data in order to extract some specific piece of information, it is
usually necessary to process the data in order to assure that it
satisfies certain assumptions implied by the method. Examples
are Re-sampling in order to assure that the image coordinate
system is correct. Noise reduction in order to assure that
sensor noise does not introduce false information. Contrast
enhancement to assure that relevant information can be
detected. Scale-space representation to enhance image
structures at locally appropriate scales.
III.3 FEATURE EXTRACTION
Image features at various levels of complexity are
extracted from the image data. Typical examples of such
features are Lines, edges and ridges. Localized interest
points such as corners, blobs or points.
III.4 DETECTION/SEGMENTATION
At some point in the processing a decision is made about
which image points or regions of the image are relevant for
further processing. Examples are Selection of a specific set of
interest points Segmentation of one or multiple image regions
which contain a specific object of interest.
III.5 HIGH-LEVEL PROCESSING
At this step the input is typically a small set of data, for
example a set of points or an image region which is assumed
to contain a specific object. The remaining processing deals
with, for example: Verification that the data satisfy modelbased
and application specific assumptions. Estimation of
application specific parameters, such as object poses or
objects size. Classifying a detected object into different
categories.
III.6. HARDWARE CONTROLLING
It is a microcontroller based device control system where
connected appliances can be operated using digital logic 1 or
0. In proposed system emergency alarm or the lighting system
can be activated.
IV. REQUIREMENT ANALYSIS
CMUcam4 v1.02 Firmware
The CMUcam4 is a fully programmable embedded
computer vision sensor. The main processor is the Parallax
P8X32A (Propeller Chip) connected to an Omni Vision 9665
CMOS camera sensor module.
IV.1 INTRODUCTION OF PARALLAX P8X32A
The design of the Propeller chip frees application
developers from common complexities of embedded systems
programming. For example: Eight processors perform
simultaneous processes independently or cooperatively,
sharing common resources through a central hub. The
Propeller application designer has full control over how and
when each cog is employed; there is no compiler-driven or
operating system-driven splitting of tasks among multiple
cogs. This method empowers the developer to deliver
absolutely deterministic timing, power consumption, and
response to the embedded application.
Asynchronous events are easier to handle than with
devices that use interrupts. The Propeller has no need for
interrupts; just assign some cogs to individual, high-bandwidth
tasks and keep other cogs free and unencumbered. The result
is a more responsive application that is easier to maintain. A
shared System Clock allows each cog to maintain the same
time reference, allowing true synchronous execution.
IV.2 PROGRAMMING ADVANTAGES
The object-based high-level Spin language is easy to learn, with special commands that allow developers to quickly exploit the Propeller chip’s unique and powerful features. Propeller Assembly instructions provide conditional execution and optional flag and result writing for each individual instruction. This makes critical, multi-decision blocks of code more consistently timed; event handlers are less prone to jitter and developers spend less time padding, or squeezing, cycles.
IV.3 APPLICATIONS
The Propeller chip is particularly useful in projects that can be vastly simplified with simultaneous processing, including: Industrial control systems, Sensor integration, signal processing, and data acquisition, Handheld portable human-interface terminals, Motor and actuator control, User interfaces requiring NTSC, PAL, or VGA output, with PS/2 keyboard and mouse input, Low-cost video game systems, Industrial, educational or personal-use robotics, Wireless video transmission (NTSC or PAL).
IV.4 PROGRAMMING PLATFORM SUPPORT
The Propeller Demo Board convenient means to test-drive the Propeller chip’s varied capabilities through a host of device Interfaces on one compact board. Main features are P8X32A-Q44 Propeller Chip, 24LC256-I/ST EEPROM for program storage, Replaceable 5.000 MHz crystal, 3.3 V and 5 V regulators with on/off switch, USB-to-serial interface for programming Communication, VGA and TV output, Stereo output with 16 ?? headphone amplifier, Electret microphone input, Two PS/2 mouse and keyboard I/O connectors, 8 LEDs (share VGA pins), Pushbutton for reset, Big ground post for scope hookup, I/O pins P0-P7 are free and brought out to header, Breadboard for custom circuits.
IV.5 PROPELLER TOOL SOFTWARE
The Propeller Tool Software is the primary development environment for Propeller programming in Spin and Assembly Language. It includes many features to facilitate organized development of object-based applications: multi-file editing, code and document comments, color-coded blocks, keyword highlighting, and multiple window and monitor support aid in rapid code development. Optional view modes allow you to quickly drill down to the information you need’by hiding comment lines, method bodies, or by showing the object’s compiled documentation only. Example objects, such as keyboard, mouse, and graphics drivers, come standard with the free Propeller Tool software.
A.PROPELLENT LIBRARY AND EXECUTABLE
The Parallax Propellent software is a Windows-based tool for compiling and downloading to the Parallax Propeller chip’without using the Propeller Tool development software. The Propellent Executable provides the ability to do things like compile Spin source, save it as a binary or EEPROM image, identify a connected Propeller chip, and download to the Propeller chip, all via simple command-line switches or drag-and-drop operations.
B.PROPELLER GCC
The Propeller GCC Compiler tool-chain is an open-source, multi-OS, and multi-lingual compiler that targets the Parallax Propeller’s unique multicore architecture. Parallax has collaborated with industry experts to develop all aspects of the tool chain, including the creation of a new development environment that simplifies writing code, compilation, and downloading to a Propeller board. Using the Large Memory Model (LMM) and Extended Memory Model (XMM) gives the developer the ability to write C or C++ programs that run faster than Spin or exceed Spin’s 32 KB program size limit, respectively. Example objects, including C objects, are available through the Propeller Object Exchange.
V. RELATED WORK
VLSI design and experimental measurements of a single-chip CMOS image sensor with moving object detection and localization capability [6].Motion events are first detected using a frame differencing scheme; then they are processed by an on-the-fly clustering processor to localize the motion objects in the scene [7].Unlike existing systems relying on external FPGAs or CPLDs to perform object localization, our system does not require any external computation or storage
[8]. The proposed algorithm is integrated on chip, featuring compact silicon implementation and little power consumption. The proposed design is an ideal candidate of wireless sensor network node, for applications such as assisted living monitors, security cameras, and even robotic vision [9]. Future improvements include adoption of a dynamic resolution pixel array and event based object tracking [10].
Fig.CMU cam4 kit interface on computer
Fig. Motion detection by CMU cam4
VI. CONCLUSION
It is possible to make a low-power camera based motion detection system with the help of CMU cam4 kit. With enough time and testing, an advanced motion detection algorithm could be designed that minimizes power consumption while detecting motion easily. More advanced cameras with higher resolutions and better processors in the future will further this effort to create a truly smart sensor that knows when a room is occupied and when a room is not occupied.