No abstract

No abstract

Weight is an important property of an object which is used to manipulate an object efficiently. Approximating weight of an object is the key area which involves many scientific calculations. However, manual calculations involve many overheads and are not reliable. Recently some computerised techniques are used but they require huge input based on assumptions. Weight estimation of Shaped Object using Image processing Techniques is a prominent solution which targets this problem. Image processing reduces this burden and can be used effectively. Initially, the shape of the object is identified by capturing the image and applying boundary detection and pattern matching. Two methodologies are implemented in classification of images, classical Pattern Matching Technique and Convolutional neural network Then by accepting the dimensions of the object and the predefined values for density of materials the weight estimation can be carried out. This paper briefly discusses the various weight estimation techniques for different shapes such as Cylinders, Cubes and Cones which is efficiently calculated after detection. This problem can be successfully applied for weight estimation of various steel castings used in automobile and oil refinery industries. Based on weight, liquid metal will be melted in furnaces for pouring into moulds.

No abstract