Image Classification | Fresco Play

Image Classification | Fresco Play

Monday, May 22, 2023
~ 6 min read
Image Classification | Fresco Play

Question 1: Identify the unstructured data from the following

Answer: Both Image and video


Question 2: Which preprocessing technique is used for dimensinality reduction?

Answer: SVD


Question 3: True Positive is when

Answer: The predicted instance and the actual instance are not negative


Question 4: True Negative is when

Answer: The predicted instance and the actual instance are negative


Question 5: Technique used to evaluate a classifier by dividing the data set into train set to train the classifier and test set to test the same.

Answer: cross validation


Question 6: Which of the following is not an example of CNN architectures?

Answer: none


Question 7: A technique used to depict the performance in a tabular form that has 2 dimensions namely “actual” and “predicted” sets of data is called ___________.

Answer: confusion matrix


Question 8: High classification accuracy always indicates a good classifier.

Answer: False


Question 9: In Supervised learning, class labels of the training samples are ___________.

Answer: known


Question 10: Classification where each data is mapped to more than one class is called ____________.

Answer: multi label classification


Question 11: Choose the correct sequence from the following

Answer: Image Analysis -> PreProcessing -> Model Building--> Predict


Question 12: The improvement of the image data that suppresses distortions or enhances image features is called ____________.

Answer: Image Pre-Processing


Question 13: SVM is a __________.

Answer: supervised learning algorithm.


Question 14: Select the correct statements about Nonlinear classification.

Answer: Kernel trick non linear


Question 15: Supervised learning differs from unsupervised learning. Supervised learning requires ____________.

Answer: labeled data


Question 16: Which of the following is not a performance evaluation measure?

Answer: Decision tree


Question 17: Which of the following is a feature extraction technique?

Answer: all


Question 18: The scale-invariant feature transform can be used to detect and describe local features in images.

Answer: True


Question 19: Which of the given hyper parameter(s), when increased may cause random forest to over fit the data?

Answer: depth of tree


Question 20: The normalized linear combination of the original predictors in a data set is called ____________.

Answer: Principal component


Question 21: TF-IDF is a common methodology used in pre-processing of images.

Answer: False


Question 22: The process of changing the pixel intensity values to achieve consistency in dynamic range for images is ___________.

Answer: Image normalisation


Question 23: Which classification techniques involves finding the eigenvalues and eigenvectors?

Answer: SVD


Question 24: What is the function that converts K-dimensional vector containing real values to the same shaped vector of real values in the range of (0,1), whose sum is 1?

Answer: softmax


Question 25: The variation present in the PCs decrease as we move from the 1st PC to the last one.

Answer: True


Question 26: Select the correct option that directly achieves multi-class classification (without support of binary classifiers).

Answer: K Nearest Neighbor


Question 27: Clustering is a supervised classification

Answer: False


Question 28: Choose the correct sequence for classifier building from the following:

Answer: Initialize -> Train - -> Predict-->Evaluate


Question 29: PCA

Answer: Principal component analysis


Question 30:  SIFT stands for

Answer: Scale Invariant Feature Transform


Question 31: Higher value of which of the following hyperparameters is better for decision tree algorithm?

Answer: Cannot say


Question 32: Netflix OSS is example

Answer: Client side


Question 33: A classifer that can compute using numeric as well as categorical values is

Answer: 1. Decision Tree Classifier 2. Naive Bayes Classifier


Question 34: Which of the following is not a characteristics of HOG?

Answer: 1. Used in sliding window fashion  and  Computer gradients 2. Compute gradients in the region are to be described


Question 35: Images, documents are examples of

Answer: unstructured


Question 36: The most widely used package for machine learning in python is

Answer: sklearn


Question 37: Pruning is a technique associated with

Answer: Decision tree


Question 38: SIFT computes the gradient histogram only for patches where as HOG is computed for an entire image.

Answer: False


Question 39: The fit(X, y) is used to ___________.

Answer: Train the classifier


Question 40: Which of the following is not a preprocessing technique used for image processing?

Answer: Noise filtering


Question 41: The first layer in a CNN is never a Convolutional Layer

Answer: False


Question 42: HOG is simplified version of SIFT

Answer: False


Question 43: Unsupervised classification identifies larger number of spectrally-distinct classes than supervised classification.

Answer: True


Question 44: Choose the right options based on Pooling.

Answer: All


Question 45: Which algorithm can be used for matching local regions in two images?

Answer: SIFT


Question 46: Which type of cross validation is used for imbalanced dataset?

Answer: Split


Question 47: Which among the following is True? A. SIFT is used for identification of specific objects B. HOG is used for classification

Answer: Both A and B


Question 48: Which one of the following is not a classification technique?

Answer: Startified shuffle split


Question 49: HOG refers to

Answer: Histogram of Oriented Gradient


Question 50: In SVD, the matrix A of dimension m x n can be decomposed in to A=USVT, where V is a ___________.

Answer: n x n orthonormal matrix


Question 51: The dimensionality reduction technique that efficiently represents interesting parts of an image as a compact feature vector.

Answer: Edge detection


Question 52: Which classifier involves finding Optimal hyperplane for linearly separable Patterns?

Answer: SVM


Question 53: GradientDescent is one of Backward propagation techniques to find the best set of parameters of the network.

Answer: True


Post a comment

Comments

Join the conversation and share your thoughts! Leave the first comment.

Get your FREE PDF on "100 Ways to Try ChatGPT Today"

Generating link, please wait for: 60 seconds

Checkout all hot deals now 🔥

Search blogs

No blog posts found