Academia.eduAcademia.edu

Figure 3 – uploaded by GRD JOURNALS

See full PDFdownloadDownload figure
Fig. 3: Transmission of wave from Ultrasonic Sensor

Figure 3 Transmission of wave from Ultrasonic Sensor

Related Figures (15)

Fig. 1: Block Diagram of proposed system
Fig. 1: Block Diagram of proposed system
Fig. 2: Working principle of Ultrasonic Sensor The Ultrasonic HC-SR04 sensor detects objects by emitting a short ultrasonic burst and then "listening" for the echo. Principle of the Ultrasonic sensor is shown in the fig.2. Under control of a host microcontroller (trigger pulse), the sensor emits a short 40 kHz (ultrasonic) burst. This burst travels through the air at about 1130 feet per second, hits an object and then bounces back to the sensor. The ultrasonic sensor provides an output pulse to the host that will terminate when the echo is detected and hence the width of this pulse corresponds to the distance to the target. This output is fed to the Raspberry pi which converts this signal into audio signal [1]. Then this audio signal reaches the user through earphones/speakers.
Fig. 2: Working principle of Ultrasonic Sensor The Ultrasonic HC-SR04 sensor detects objects by emitting a short ultrasonic burst and then "listening" for the echo. Principle of the Ultrasonic sensor is shown in the fig.2. Under control of a host microcontroller (trigger pulse), the sensor emits a short 40 kHz (ultrasonic) burst. This burst travels through the air at about 1130 feet per second, hits an object and then bounces back to the sensor. The ultrasonic sensor provides an output pulse to the host that will terminate when the echo is detected and hence the width of this pulse corresponds to the distance to the target. This output is fed to the Raspberry pi which converts this signal into audio signal [1]. Then this audio signal reaches the user through earphones/speakers.
Fig. 4: Reception of echo by Ultrasonic Sensor
Fig. 4: Reception of echo by Ultrasonic Sensor
Fig. 5: Flow chart for obstacle detection
Fig. 5: Flow chart for obstacle detection
wee Ne ee ae ee, Ne ee It is a machine learning based approach where a cascade function is trained from a lot of positive and negative images. Then need to extract features from it. Haar features are shown in the fig.6. They are just like convolutional kernel. Each feature is single value obtained by subtracting sum of pixels under white rectangle from sum of pixels under black rectangle. Among all these features calculated, most of them are irrelevant [3]. This can be reduced by using Ada boost classifier. Ada boost reduces the redundant features. Here instead of summing up all the pixels the integral image is used as shown in fig.7.
wee Ne ee ae ee, Ne ee It is a machine learning based approach where a cascade function is trained from a lot of positive and negative images. Then need to extract features from it. Haar features are shown in the fig.6. They are just like convolutional kernel. Each feature is single value obtained by subtracting sum of pixels under white rectangle from sum of pixels under black rectangle. Among all these features calculated, most of them are irrelevant [3]. This can be reduced by using Ada boost classifier. Ada boost reduces the redundant features. Here instead of summing up all the pixels the integral image is used as shown in fig.7.
To get a new pixel value the top pixels and left pixels are added then all the values around the patch are added to obtain the sum of all pixel value. Ada boost determines relevant features and irrelevant features. A fter identifying relevant features and irrelevant features the Ada boost assigns a weight to all of them [4].
To get a new pixel value the top pixels and left pixels are added then all the values around the patch are added to obtain the sum of all pixel value. Ada boost determines relevant features and irrelevant features. A fter identifying relevant features and irrelevant features the Ada boost assigns a weight to all of them [4].
~ As the function of natural neural network system of brain, the computer created artificial network has inputs for insertin he data, outputs for providing the network output and hidden layer for processing the data and training of the network as show n Fig.8. The number of neurons in the input layer is equal to number of Eigen faces, number of neurons in the hidden layer is 1( nd the type is Feed forward back propagation network.
~ As the function of natural neural network system of brain, the computer created artificial network has inputs for insertin he data, outputs for providing the network output and hidden layer for processing the data and training of the network as show n Fig.8. The number of neurons in the input layer is equal to number of Eigen faces, number of neurons in the hidden layer is 1( nd the type is Feed forward back propagation network.
Fig. 9: Flow Chart Ne VYILIULIGLUIL UL INGULGIL INOLUWUL RO Face descriptors of the test image calculated from the Eigen faces are applied as input to all the networks and they are simulated Simulated results are compared and a maximum output greater than a predefined threshold level confirms that the test image belongs to the recognized person with the maximum output [5].
Fig. 9: Flow Chart Ne VYILIULIGLUIL UL INGULGIL INOLUWUL RO Face descriptors of the test image calculated from the Eigen faces are applied as input to all the networks and they are simulated Simulated results are compared and a maximum output greater than a predefined threshold level confirms that the test image belongs to the recognized person with the maximum output [5].
Fig. 10: Flow chart for image to speech conversion bityttoll OVUUULL. LL UG BG ULLLOLUIIL OY LLU OLS LOU LEU UU UVULE OU UE 21 ULI Io, GU OUULIUDS YULLY ULLLULUIIL, It can run as a command line program to speak text from a file or from standard input. It includes different voices, whose characteristics can be altered. It can speak text, or produce speech output as a WAV file. SSML (Speech Synthesis Markup Language) is supported (not complete), and also HTML. The program and its data, including several languages, total about 1 Mbyte. It can translate text to phoneme codes. It can be used as a front end for Mbrola voices and perhaps other speech synthesis engines. It can be written in C++.
Fig. 10: Flow chart for image to speech conversion bityttoll OVUUULL. LL UG BG ULLLOLUIIL OY LLU OLS LOU LEU UU UVULE OU UE 21 ULI Io, GU OUULIUDS YULLY ULLLULUIIL, It can run as a command line program to speak text from a file or from standard input. It includes different voices, whose characteristics can be altered. It can speak text, or produce speech output as a WAV file. SSML (Speech Synthesis Markup Language) is supported (not complete), and also HTML. The program and its data, including several languages, total about 1 Mbyte. It can translate text to phoneme codes. It can be used as a front end for Mbrola voices and perhaps other speech synthesis engines. It can be written in C++.
Fig.12 Capturing of face by webcam
Fig.12 Capturing of face by webcam
Fig.16 Detected image Fig.15 capturing the text image
Fig.16 Detected image Fig.15 capturing the text image
Fig.19 Voice output through ear phones Fig.18 Capturing of currency
Fig.19 Voice output through ear phones Fig.18 Capturing of currency
Related topics:
Image ProcessingObstacle Detection
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved