• Vishwanath Burkpalli

This paper presents a comparative study of certainon segmentation techniquees for classification of food objects from multiple food objects’ images. We have used color and texture features for identification and the algorithms, namely, thresholding, region growing, k-means clustering and Chan-Vase are being adopted for segmentation. An active contour method based on Chan-Vase model has given good segmentation results for multiple food objects. A back propagation neural network (BPNN) is used as the classifier. Different types of south Indian food objects like Idli, Vada, Puri, Cutlet, and Samosa, are considered for the study. The classification accuracy for combined features is found to be 82%. The work finds application in automatic serving by robots in pharamaceutical industry, food industry, hotels, motels, Cafeteria, shaping malls, etc.

Image Segmentation, Color Features, Textural Features, Back Propagation Neural Network

"Comparative Study of Certain Segmentation Techniques for Classification of Food Objects’ Images", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 7, page no.734-743, July-2018, Available :http://www.jetir.org/papers/JETIR1807119.pdf

"Comparative Study of Certain Segmentation Techniques for Classification of Food Objects’ Images", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 7, page no. pp734-743, July-2018, Available at : http://www.jetir.org/papers/JETIR1807119.pdf