EMT Practice Test

1. Question Content...

---

Question1: By visualizing the feature map of the first layer of a convolutional neural network, you can see clear target features.

Question2: Modifying the H component of the HSV color space will change the image?

Question3: TF-IDF is a calculation method based on statistics, commonly used to evaluate the importance of a word in a document set to all documents.

Question4: In the GMM-HMM model, GMM is mainly used to find the probability of a certain factor, and HMM is mainly used to model each text-speech pair.

Question5: (Judgment) The background difference method is to use the current frame in the image sequence and the background reference model for subtraction, which can realize the detection of moving objects in the image.

Question6: What is the spatial adjacency of the two pixels with coordinates (21, 13) and coordinates (22.12)?

Question7: What are the processing procedures included in image digitization? (Multiple choices)

Question8: What is the correct description about the Gaussian distribution curve?

Question9: When acquiring digital images, what factors determine the image quality of the image depends on the sensor?

Question10: In the Gaussian mixture model, since the distribution of each sample point is unknown, it is impossible to use the maximum function to solve the parameters.

Question11: The 19 in VGG19 represents the subordinate sum of the number of layers in the network?

Question12: When using gensim to implement Doc2vce, which parameter represents the specific model selected?

Question13: The histogram operation of the image is to count the distribution frequency of the gray value of the whole image. When the illumination of the image is uneven, the operation effect of the histogram will be affected, and it is necessary to use an improved algorithm for image processing.

Question14: Using binary digital segmentation, because of noise, the connected domains of all characters are glued together. What method can be used to segment characters?

Question15: The various indicators of the bilinear interpolation method are better than the nearest neighbor interpolation method. The nearest neighbor interpolation method has been eliminated by the bilinear interpolation method in practical applications.

Question16: The image filter operation is also called template operation. Common template operations include template convolution and template row.

Question17: Jieba framework is a toolkit for English character processing.

Question18: What is not an acoustic feature of speech?

Question19: What is the description of the standard normal distribution?

Question20: The request/response of HUAWEI CAI API in accordance with the design of RESTful API includes the following.

Question21: The target in the image is 15 pixels to the right relative to the center position. Which of the following shift amounts can be used to shift the translation coordinates to move the target to the center of the screen?

Question22: Adaboost is a classic parallel algorithm in the integrated algorithm. It uses multiple decision tree classifiers to betray the problem and vote for the final result.

Question23: The semantic content of the image may be very rich, and image content detection can be more intelligent.
Ready to understand image content. Generate multiple description tags from the content of the image, which scene can be applied to?

Question24: Which preset models does ModelArts currently support? (multiple choices)

Question25: In the gensim package, the parameter size of word2vce represents the dimension of the word vector.

Question26: Histogram equalization can enhance the visual effect of the image. This technique can automatically calculate the transformation function without artificially setting parameters. The operation is simple and suitable for all situations.

Question27: What is a voice processing task?

Question28: Which AI scenarios are Huawei Ascend series chips for?

Question29: Huawei Cloud EI Enterprise Intelligence provides a variety of convenient and efficient services. What is the correct process for users to use the services?

Question30: (Judgment) The Gaussian model is to accurately quantify things with a Gaussian density function, and decompose a thing into several models based on Bernoulli distribution.

Question31: In the vehicle detection scene of traffic law enforcement, the penalties for vehicles that violate regulations are rather missed, instead of false alarms, so as not to produce false fines, which of the system's instructions are high?

Question32: In order to make the feature map consistent with the input image, you can use the filling method to add an all 0 and a frame to the input image and then calculate.

Question33: (Judgment) The original value of the recording pixel is S, The pixel value after the change is D.
The mapping function of gray scale change is d=ks. When k>1, gray scale stretching of the dark part of the image can be achieved.

Question34: What is incorrect about the following analysis?

Question35: Using two 3*3 convolutions to check a three-channel color image into a convolution, how many channels are there for the feature map?

Question36: The convergence point of gradient descent calculation depends on the cost function.

Question37: Which of the following scenarios will use target tracking technology?

Question38: Named entity recognition refers to recognizing entities with specific meaning in text.

Question39: Many mobile phones provide eye protection screen display mode, which can reduce blue light and relieve fatigue. This function can be achieved by adjusting the displayed image.

Question40: Named entity recognition refers to recognizing entities with specific meanings in the text. Which ones are mainly included? (Multiple Choice)

Question41: Inverting the RGB channels of a color image separately will change the brightness of the image, but will not change the color of the image.

Question42: In the Huawei cloud content detection service, which of the following parameters is not a public request message header.

Question43: Image target detection algorithm needs to be completed?

Question44: (Judgment) In general, Huawei's artificial intelligence development strategy is based on continuous investment in basic research and AI talent training, creating a full-stack full-scenario AI solution and an open global ecology.

Question45: In the Huawei Cloud Face Recognition Service, users can create multiple face libraries without restrictions.

Question46: Overall, Huawei's development strategy for artificial intelligence is based on continuous investment in basic research and AI training to create a full-stack, full-scenario AI solution and an open global ecosystem.

Question47: What algorithm engines are supported by ModelArts training assignments? (Multiple Choice)

Question48: When the parameter sg=1 of word2vce in the gensim package represents the selection of the CBOW algorithm.

Question49: What is linguistic content?

Question50: Which of the following is not a common method of text vectorization?

Question51: What are the difficulties of named entity recognition?

Question52: When taking light, the face is relatively dark. Use gamma correction to enhance the image to achieve a clear light effect. Which of the Y limit can be selected. (Single choice)

Question53: What types of phonetics include?

Question54: What is the correct description of the nature of the probability density function?

Question55: The OTSU algorithm traverses the possible thresholds, and selects the segmentation threshold of the maximum value in the foreground and background regions as the optimal threshold.

Question56: The semantic content of the image may be very rich, and image content detection can be more intelligent.
Ready to understand image content. Generate multiple description tags from the content of the image, which scene can be applied to?

Question57: Speech processing is a general term for studying various speech technologies such as speech utterance process, speech statistical characteristics, language perception, etc.

Question58: What is the most accurate estimation process?
1. Log-likelihood function
2. Probability density function
3. Likelihood function
4. Derivative and make the equation 0
5. Solving equations

Question59: Speech recognition technology is to let the machine convert speech into text through recognition and understanding.

Question60: Since the current speech processing technology is based on digital computing, it is also called digital speech signal processing.

Question61: Which of the following steps in the MFC extraction process?

Question62: Maximum estimation requires knowing the distribution type of sample data.