In this work, the Relevant title was given for the proposed work. Different techniques have been mentioned to retrieve the image from data sets. Also discussed about the challenges in retrieving the image, but the parameters are involved in the challenges are not discussed. some more clarity/explanation is required in decreasing the volume of image and the size of the image. In the proposed work they had discussed only about the grayscale image. The content provided is not enough for the feature percentage explanation. Under the result discussion, the comparison is made between the different techniques, but not the features.One part of the work discussed in the result.

In the introduction, they have discussed regarding the Content-based image retrieval technique to retrieve the image from the image collection. Due to the increase in the use of the digital camera and social media tends to increase image collection. Traditional image retrieval technique also discussed, but the drawbacks are not explained clearly. The challenges in the proposed work is discussed. The low level features and high level features are also discussed.

In this proposed methodology, visual word fusion or integration is discussed. Different fusion techniques are considered in image retrieval. Training sets and K means clustering algorithm used in the retrieving process. Dictionary size and image size are discussed. And also the feature percentage of the image retrieval is not mentioned clearly.

Performance comparison of standalone SRUF, standalone FREAK, and features a fusion of SURF - FREAK techniques on different sizes of the dictionary for the core 1500 are discussed.Visual word fusion technique based on different sizes are considered.The image retrieval performance measure such as precision and recall are also compared with state of art CBIR technique.

Regarding the combination of the techniques, only comparison is made between the different content-based image retrieval techniques. The implementation of the method is not explained clearly, in the parameters and feature percentage. Also only the gray scale image is considered.