spectrometer.py

# Based on: https://github.com/uutzinger/C12880MA

import serial
import serial.tools.list_ports
import time
import matplotlib.pyplot as plt
import numpy as np
import scipy.signal as signal

def find_serial_device(vid=None,pid=None,product=None):
    devices = []
    for port in serial.tools.list_ports.comports():
#        print(port.device,port.vid, port.pid, port.product)
        if (vid != None) and (vid != port.vid):
            continue
            
        if (pid != None) and (pid != port.pid):
            continue

        if (product != None) and (product != port.product):
            continue
        
        devices.append(port.device)
            
    return devices

class MicroSpec(object):
    def __init__(self, port):
        self._ser = serial.Serial(port,baudrate=115200)
    def set_integration_time(self, seconds):
        cmd = "SPEC.INTEG %0.6f\n" % seconds
        self._ser.write(cmd.encode('utf8'))
    def read(self):
        self._ser.write(b"SPEC.READ?\n")
        sdata = self._ser.readline()
        sdata = np.array([int(p) for p in sdata.split(b",")])
        self._ser.write(b"SPEC.TIMING?\n")
        tdata = self._ser.readline()
        tdata = np.array([int(p) for p in tdata.split(b",")])
        return (sdata, tdata)
    def read_many(self, integration_time, count):
        spec.set_integration_time(integration_time)
        stotal = None
        for sample in range(0,count):
            sdata, tdata = self.read()
            if type(stotal) == type(None):
                stotal = sdata
            else:
                stotal += sdata
        
        return stotal/sample_count
        
    
cal_data = {
    "A_0" : 3.062951791e2,
    "B_1" : 2.720613040e0,
    "B_2" : -1.302773396e-3,
    "B_3" : -7.101037823e-6,
    "B_4" : 8.279199710e-9,
    "B_5" : 5.742926217e-12
}

def correct(pix, cal_data):
    return (cal_data["A_0"]) \
            + (cal_data["B_1"]*pix) \
            + (cal_data["B_2"]*(pix**2)) \
            + (cal_data["B_3"]*(pix**3)) \
            + (cal_data["B_4"]*(pix**4)) \
            + (cal_data["B_5"]*(pix**5))

# frequency points
nm = [correct(i, cal_data) for i in range(1,289)]
    
spec = MicroSpec(find_serial_device(vid=5824, pid=1155)[0]) # Teensy 3.2

# integration_time = 0.000085 # read time, in seconds
# sample_count = 20  # number of samples to average
integration_time = 0.000085 # read time, in seconds
sample_count = 20  # number of samples to average

datas = {}
datas['dark'] = spec.read_many(integration_time, sample_count)

def normalize(a):
    b=np.ndarray.sum(a,axis=0)
    return a/b

integration_time = 0.00000085 # read time, in seconds
datas['uv'] = spec.read_many(integration_time, sample_count)

plt.rcParams['figure.figsize'] = [20, 10]
for label, spectrum in datas.items():
    if label=='dark' or label=='warm' or label=='natural':
         continue
    d = normalize(spectrum-datas['dark'])
    plt.plot(nm,d,label=label)
    
    peaks, _ = signal.find_peaks(d, distance=20, height=.005)
    np.diff(peaks)
    
    v = np.array(nm)
    plt.plot(v[peaks], d[peaks], "x")
    print(label, "peaks(nm):", [int(i) for i in v[peaks]])

    #plt.show()
    

#plt.plot(nm,(datas['cool']-datas['dark'])*.65+(datas['warm']-datas['dark'])*.50,label='cool*.65+warm*.5')
    
plt.title('Normalized Spectrum')
plt.xlabel('Wavelength(nm)')
plt.ylabel('%')
plt.legend()
        
#data['spectrums'].append(stotal/sample_count)