Scribd
Upload
8 views

Lighting Consistency Analysis

Uploaded by

shravani26127
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
8 views

Lighting Consistency Analysis

Uploaded by

shravani26127
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 9

Lighting Consistency Analysis

Solution Approach
STEP 1 . COLLECT DATA SET AND ANNOTATE THE IMAGES
STEP 2 : DOWNLOAD THE DATASET IN SUITABLE FORMAT

STEP 3 : USE DATA FOR PREPROCESSING

STEP 4 : FEATURE EXTRACTION

FEATURE EXTRACTION:

CONVOLUTIONAL LAYERS:

EXTRACT FEATURES FROM THE IMAGE USING FILTERS THAT DETECT PATTERNS

RELATED TO LIGHTING CONDITIONS. THESE PATTERNS MIGHT INCLUDE SHADOWS,

HIGHLIGHTS, AND CONTRAST VARIATIONS.

POOLING LAYERS: REDUCE THE SPATIAL DIMENSIONS OF THE FEATURE MAPS WHILE

PRESERVING ESSENTIAL INFORMATION


FEATURE CLASSIFICATION:

•FULLY CONNECTED LAYERS:

FLATTEN THE FEATURE MAPS AND CONNECT THEM TO FULLY CONNECTED


LAYERS TO CLASSIFY THE EXTRACTED FEATURES.

•OUTPUT LAYER:

•THE FINAL LAYER WITH NEURONS CORRESPONDING TO DIFFERENT LIGHTING


CONDITIONS (E.G., "BRIGHT," "DARK," "NATURAL," "ARTIFICIAL").
TRAINING THE CNN:

LOSS FUNCTION: CHOOSE A SUITABLE LOSS FUNCTION, SUCH AS


CATEGORICAL CROSS-ENTROPY, TO MEASURE THE DIFFERENCE BETWEEN
THE PREDICTED AND ACTUAL LIGHTING CONDITIONS

TRANSFER LEARNING:

WE ALSO USE A PRE-TRAINED CNN (E.G., VGG, RESNET) AS A STARTING POINT


AND FINE-TUNE IT FOR YOUR SPECIFIC TASK.
TRANSFER LEARNING:

CHOOSE A PRE-TRAINED MODEL:


RESNET, VGG, INCEPTION, MOBILENET

You might also like