1. 研究目的与意义（文献综述）

imageclassification using deep learning

imageclassification

imageclassification refers to the task of extracting information classes from a multibandraster image. the resulting raster from image classification can be used tocreate thematic maps. depending on the interaction between the analyst and thecomputer during classification, there are two types of classification:supervised and unsupervised.

2. 研究的基本内容与方案

基本内容：

研究和学习卷积神经网络，建立一个神经网络，根据现有一系列图片，训练系统具有分辨图像类别的能力。首先学习和分析现有相关开源系统，梳理出使用的分类和识别图像的方法，以及熟悉编程方法；然后基于研究和总结，在分辨准确度、处理速度、训练样本集需求等方面提出一套有效改进的方案并实现验证。

3. 研究计划与安排

2019.02.26至2019.03.08 查阅相关资料，完成开题报告工作

2018.03.09至2019.03.31 数据收集和编码

4. 参考文献（12篇以上）

[1] abuzaghleh, b. d. barkana, and m.faezipour, “noninvasive real-time automated skin lesion analysis system formelanoma early detection and prevention,” ieee journal of translationalengineering in health and medicine, vol. 3, pp. 1–12, 2015.

[2] the ham10000 dataset, a largecollection of multi-source dermatoscopic images of common pigmented skinlesions

[3] visin, francesco, et al."renet: a recurrent neural network based alternative to convolutionalnetworks." arxiv preprint arxiv:1505.00393 (2015).