Student Projects

VE/VM450

Compact Platform for Lensless Inline Holographic Microscope

Instructors: Prof. Jigang Wu
Team Members: Yibo Chen, Mingliang Duanmu, Xincheng Wu, Jiawen Xu, Yang Yu

Project Video

Team Members

Team Members:

Yibo Chen, Mingliang Duanmu, Xincheng Wu, Jiawen Xu, Yang Yu

Instructors:

Prof. Jigang Wu

Project Description

Problem

Sometimes we need a microscope in various environments, especially for medical or learning usage. However, a traditional microscope may not be a good choice due to its large scale and inconvenience.
Our group’s goal is to design a compact platform for lensless inline holographic microscope including robust hardware and efficient software, which is affordable and user-friendly. What is more, it should be portable that the user can bring it everywhere.

Concept Generation

Raw data will be transferred from the CMOS imaging sensor to the image processing program. Later, a bitmap image file will be given to the server and an algorithm will be applied to reconstruct the object image from its hologram. Then, users can interact with the server to get the reconstructed and high-resolution image on their own device.
Fig.1 Flow Chart

Design Description

The design includes both hardware and software parts.
For the hardware part, it can be divided into 3 parts. The top part is designed to place the CMOS imaging sensor and the microscope slide. The middle part is a sleeve for adjusting the distance from light source and the CMOS imaging sensor. The bottom part is to place the battery box.
Fig.2 The whole hardware design
For the software part, it is divided into 2 parts, the frontend and the backend. The frontend mainly contains a user interface in the form of a website. It has a dashboard that provides access to different functions like presenting the project details and displaying the result of the microscope. The backend is deployed on our cloud server, which consists of a file management system and the main image processing program based on our algorithm.
Fig.3 The whole software design
Our software system cooperates with our hardware to form a highly efficient data flow that allows users to check for the high-resolution microscope results conveniently within a short time.

Modeling and Analysis

A C++ model is built to reconstruct the object image from its hologram. It is based on scalar diffraction theory. The model analyzes the data in frequency domain and manipulates the data by applying a phase shift corresponding to each space frequency.
Fig.4 Reconstructed Image

Validation

Validation Process:
For size, a ruler can be used to measure the length, height and width of the product easily.
For resolution and pixel size, check the property of the figure the sensor captures.
For image processing speed, a timer can be used to measure the time to process and display the image on the screen.
Some other specifications can also be verified via easy experiments.
Validation Results:
According to validation part, most specifications can be met.
√ Pixel size <= 3.2 μm2 √ Resolution >= 1500*2000
√ Size <= 100*100*150mm
√ Cost <= 700RMB
• Time for image processing<= 5s
√ means having been verified and · means to be determined.

Conclusion

Using our hardware and software, the users can observe micrometer-level things easily. After transferring the raw data from the CMOS imaging sensor of the hardware to the cloud server, a reconstructed image will be generated and stored. Then users can visit the server at anytime to get the high-resolution images on their own device.

Acknowledgement

Sponsor: Jigang Wu from UM-SJTU Joint Institute
UM-SJTU JOINT INSTITUTE