Morphologically derivative languages form words by fusing stems and suffixes, stems are important to be extracted in order to make cross lingual alignment and knowledge transfer. As there are phonetic harmony and disharmony when linguistic particles are combined, both phonetic and morphological changes need to be analyzed. This paper proposes a multilingual stemming method that learns morpho-phonetic changes automatically based on character based embedding and sequential modeling. Firstly, the character feature embedding at the sentence level is used as input, and the BiLSTM model is used to obtain the forward and reverse context sequence, and the attention mechanism is added to this model for weight learning, and the global feature information is extracted to capture the stem and affix boundaries; finally CRF model is used to learn more information from sequence features to describe context information more effectively. In order to verify the effectiveness of the above model, the model in this paper is compared with the traditional model on two different data sets of three derivative languages: Uyghur, Kazakh and Kirghiz. The experimental results show that the model in this paper has the best stemming effect on multilingual sentence-level datasets, which leads to more effective stemming. In addition, the proposed model outperforms other traditional models, and fully consider the data characteristics, and has certain advantages with less human intervention.

Aiming at the problems of low brightness and information distortion in most image dehazing algorithms, an image dehazing algorithm based on morphological and exposure enhancement was proposed. An initial transmittance is constructed based on dark channel prior, and through a linear model that can estimate haze density, then the transmittance is optimized by morphological closing and opening operations and gradient domain guided filtering. Atmospheric light is obtained by using minimum filter and quad-decomposition algorithm. An image compensation method is proposed to enhance the details of the dehazed image, an exposure enhancement method is used to improve the brightness, and finally converted to the YUV color space and the Y channel is denoised using the BM3D algorithm.

Uyghur and Kazakh are typical agglutinative languages. Semantic and syntactic lexicons are derived from stems or roots by adding affixes. A keyword in a sentence can be the notional particles which are the semantic stems. The stems in derivative lexicons are deformed or harmonized with annexed affixes, so a special stem extraction process is necessary. This paper proposes a keyword extraction method based on a neural network stemming model. In order to verify the effectiveness of stem unit, keyword extraction is carried out on both word and stem units respectively. The experimental results show that the effects of keywords extraction through the above two particles are in obvious contrast. The keyword extracted by stemming has the best results, and the quality and efficiency of the extracted keywords are greatly improved.