Banana cultivation is one of the main agricultural elements in India, while the common problem of cultivation is that the crop has been influenced by several diseases, while the pest indications have been needed for discovering the infections initially for avoiding the financial loss to the farmers. This problem will affect the entire banana productivity and directly affects the economy of the country. A hybrid convolution neural network (CNN) enabled banana disease detection, and the classification is proposed to overcome these issues guide the farmers through enabling fertilizers that have to be utilized for avoiding the disease in the initial stages, and the proposed technique shows 99% of accuracy that is compared with the related deep learning techniques.

Discrete Wavelet Transform is an effective and important process in digital image processing. The Discrete Wavelet Transform is used for significant compression, segmentation, classification, image enhancement and image fusion of web images that are to be viewed in short page loading time with bandwidth as major constraint. The DWT provides spatial frequency information and the spatial location. The main advantage of this nature is that concurrent information helps us to reduce redundancy and increases the bandwidth more efficiently. The spatial locations of the image may or may not contain variations and at certain times may be constants also. A signal which is constant does not carry significant information. If the coefficients of constant signal are encoded along with other spatial location coefficients it is memory wastage. Discrete Wavelet Transform if applied to the image as a whole results in better frequency resolution and good spatial resolution. But still this spatial resolution is not that good. This work aims in improving the spatial resolution further. The image is segmented in space into small sub-images and Discrete Wavelet Transform is applied recursively to each and every sub image. Since the target image is spatially small in resolution as well, the same operation can be achieved with smaller Discrete Wavelet Transform Unit. Since there are only few pixels at any given processing time is faster and the bandwidth, throughput is high. Because of large throughput the entire operation can be pipelined and done in a serial manner. The proposed work reduces the number of DWT unit required for the process.

No abstract

cDNAs encoding for five mAChR subtypes (M1-M5) were cloned under different promoters in various eukaryotic vectors and each subtype was expressed in different mammalian cell lines. CHO-K1 cell line was the best for generating stable cell lines expressing muscarinic receptors. Immunofluorescence and flow cytometry revealed that expression of M1-M5 was primarily localized on the cell membrane. Western blotting and radio-ligand binding studies revealed that expression of each receptor was stable at higher passages.

No abstract