In manufacturing industry, machine vision is very important nowadays. Computer vision has been developed widely in manufacturing for accurate automated inspection. A model of automated inspection system is presented in this conceptual paper. Image processing is used for inspection of part.
It is assumed that the part after going through many previous operations comes to inspection system where the weight of the part as well as geometry made on that part is detected and later decided whether it is to be accepted or rejected with the help of image processing technique. Using MATLAB software a program is developed and pattern or geometry is detected.
REVIEW OF PAPERS
The manual activity of inspection could be subjective and highly dependent on the experience of human inspectors. So image analysis techniques are being increasingly used to automate industrial inspection. The machine vision system for automatic inspection of defects in textured surfaces has been developed.
It aimed to solve the problem of detecting small surface defects which appear as local anomalies embedded in a homogeneous texture of textile fabrics and machined surfaces. Computer vision has been for detection of defective packaging of tins of cigarettes. A large amount of research has been carried out on automated inspection of tile surfaces biscuit bake color, the color of potato chips, textile fabrics, food products and wood.
Rotary Disc Type Mechanism:
In this mechanism, the part is kept over conveyor with a noverhead gantry. Camera is fitted on the gantry exactly over the part. Conveyor is used to transfer the part from one end to another end. As soon as the part is kept over the conveyor, the camera takes image of the part and matches with parameters of the program fed beforehand.
Pusher Type Mechanism:
In pusher type mechanism, pusher is used to send the part to the acceptance centre or rejection centre. There is a rectangular block on which weighing machine is kept at the centre. The part is kept on the weighing machine. Camera is fitted on the weighing machine in such a way that camera comes exactly over the part. Weight is measured and other dimensions along with the weight are matched with the fed program.
Flip drop Mechanism is the best suited design for the project as it requires
- less space
- less moving parts
- no gantry arm for placing the part in the respective centre.
The disadvantages of other models are: All other conceptual designs used gantry arm to place the part in the respective centre, conveyors were used which complicated the design. More floor space. Use of gantry arm will increase the computational time during processing. For weighing the Aluminium block, load cell is used.
Fundamental Steps in Image Processing:
- Image acquisition: This is the first step of image processing. In this a digital image is acquired
- Image preprocessing: In the second step image is improved in a way that increase the chances for success of the other processes.
- Image segmentation: It partitions an input image into its constituent parts or objects.
- Image representation: It converts the input data to a form suitable for computer processing.
- Image description: In this step features are extracted that result in some quantitative information of interest or features that are basic for differentiating one class of objects from another.
- Image recognition: It assigns a label to an object based on the information provided by its descriptors.
- Image interpretation: A meaning is assigned to an ensemble of recognized objects in this step.
In this paper,the proposed design is being developed. This design is advantageous than other papers as this is a portable type of machine which has acceptance and rejection centre placed in a single machine. Different types of geometries can be used for image processing. This design is more advantageous for small scale industries. Different types of models or objects can be used as a work piece for testing.
Source: Yeshwantrao College Of Engg
Authors: Ms. Shubhada.K. Nagrale | Mr. S.T.Bagde