• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1# Copyright 2013 The Android Open Source Project
2#
3# Licensed under the Apache License, Version 2.0 (the "License");
4# you may not use this file except in compliance with the License.
5# You may obtain a copy of the License at
6#
7#      http://www.apache.org/licenses/LICENSE-2.0
8#
9# Unless required by applicable law or agreed to in writing, software
10# distributed under the License is distributed on an "AS IS" BASIS,
11# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12# See the License for the specific language governing permissions and
13# limitations under the License.
14
15import its.image
16import its.caps
17import its.device
18import its.objects
19import its.target
20import pylab
21import numpy
22import os.path
23import matplotlib
24import matplotlib.pyplot
25
26def main():
27    """Test that a constant exposure is seen as ISO and exposure time vary.
28
29    Take a series of shots that have ISO and exposure time chosen to balance
30    each other; result should be the same brightness, but over the sequence
31    the images should get noisier.
32    """
33    NAME = os.path.basename(__file__).split(".")[0]
34
35    THRESHOLD_MAX_OUTLIER_DIFF = 0.1
36    THRESHOLD_MIN_LEVEL = 0.1
37    THRESHOLD_MAX_LEVEL = 0.9
38    THRESHOLD_MAX_LEVEL_DIFF = 0.045
39    THRESHOLD_MAX_LEVEL_DIFF_WIDE_RANGE = 0.06
40    THRESHOLD_ROUND_DOWN_GAIN = 0.1
41    THRESHOLD_ROUND_DOWN_EXP = 0.05
42
43    mults = []
44    r_means = []
45    g_means = []
46    b_means = []
47    threshold_max_level_diff = THRESHOLD_MAX_LEVEL_DIFF
48
49    with its.device.ItsSession() as cam:
50        props = cam.get_camera_properties()
51        its.caps.skip_unless(its.caps.compute_target_exposure(props) and
52                             its.caps.per_frame_control(props))
53
54        e,s = its.target.get_target_exposure_combos(cam)["minSensitivity"]
55        s_e_product = s*e
56        expt_range = props['android.sensor.info.exposureTimeRange']
57        sens_range = props['android.sensor.info.sensitivityRange']
58
59        m = 1.0
60        while s*m < sens_range[1] and e/m > expt_range[0]:
61            mults.append(m)
62            s_test = round(s*m)
63            e_test = s_e_product / s_test
64            print "Testing s:", s_test, "e:", e_test
65            req = its.objects.manual_capture_request(
66                    s_test, e_test, 0.0, True, props)
67            cap = cam.do_capture(req)
68            s_res = cap["metadata"]["android.sensor.sensitivity"]
69            e_res = cap["metadata"]["android.sensor.exposureTime"]
70            assert(0 <= s_test - s_res < s_test * THRESHOLD_ROUND_DOWN_GAIN)
71            assert(0 <= e_test - e_res < e_test * THRESHOLD_ROUND_DOWN_EXP)
72            s_e_product_res = s_res * e_res
73            request_result_ratio = s_e_product / s_e_product_res
74            print "Capture result s:", s_test, "e:", e_test
75            img = its.image.convert_capture_to_rgb_image(cap)
76            its.image.write_image(img, "%s_mult=%3.2f.jpg" % (NAME, m))
77            tile = its.image.get_image_patch(img, 0.45, 0.45, 0.1, 0.1)
78            rgb_means = its.image.compute_image_means(tile)
79            # Adjust for the difference between request and result
80            r_means.append(rgb_means[0] * request_result_ratio)
81            g_means.append(rgb_means[1] * request_result_ratio)
82            b_means.append(rgb_means[2] * request_result_ratio)
83            # Test 3 steps per 2x gain
84            m = m * pow(2, 1.0 / 3)
85
86        # Allow more threshold for devices with wider exposure range
87        if m >= 64.0:
88            threshold_max_level_diff = THRESHOLD_MAX_LEVEL_DIFF_WIDE_RANGE
89
90    # Draw a plot.
91    pylab.plot(mults, r_means, 'r.-')
92    pylab.plot(mults, g_means, 'g.-')
93    pylab.plot(mults, b_means, 'b.-')
94    pylab.ylim([0,1])
95    matplotlib.pyplot.savefig("%s_plot_means.png" % (NAME))
96
97    # Check for linearity. Verify sample pixel mean values are close to each
98    # other. Also ensure that the images aren't clamped to 0 or 1
99    # (which would make them look like flat lines).
100    for chan in xrange(3):
101        values = [r_means, g_means, b_means][chan]
102        m, b = numpy.polyfit(mults, values, 1).tolist()
103        max_val = max(values)
104        min_val = min(values)
105        max_diff = max_val - min_val
106        print "Channel %d line fit (y = mx+b): m = %f, b = %f" % (chan, m, b)
107        print "Channel max %f min %f diff %f" % (max_val, min_val, max_diff)
108        assert(max_diff < threshold_max_level_diff)
109        assert(b > THRESHOLD_MIN_LEVEL and b < THRESHOLD_MAX_LEVEL)
110        for v in values:
111            assert(v > THRESHOLD_MIN_LEVEL and v < THRESHOLD_MAX_LEVEL)
112            assert(abs(v - b) < THRESHOLD_MAX_OUTLIER_DIFF)
113
114if __name__ == '__main__':
115    main()
116